Harmless Soul

Author(s): W J Johnson, Dayanand Bhargav
Publisher: Motilal Banarasidas
Catalog link: https://jainqq.org/explore/001196/1

Page #1

Lala S. L. Jain Research Series
pe
HARMLESS
SOULS
Karmic Bondage and Religious Change in Early Jainism with Special Reference to Umāsvāti and Kundakunda
B0022
W. J. JOHNSON

Page #2

T Tarmless Souls considers doctrinal
change and its implication for 11 religious practice in early Jainism. It focuses on evolving views about the conditions necessary for karmic bondage and the liberation of embodied souls, as described in parts of the early Svetāmbara canon, Umāsvāti's Tattvārtha Sūtra, and the major works of Kundakunda.
Particular attention is paid to the roles that knowledge and selfrealisation play on Kundakunda's path to liberation. The strategies used to preserve ascetic discipline when faced by the challenge of this emphasis on gnosis are explored in the light of various interpretations of Kundakunda's 'two truths' doctrine.
It is concluded that, although significant doctrinal changes do take place, these are not permitted to affect Jaina practice, for it is just such ascetic practice which constitutes and preserves the distinctive religious and social identity of the Jaina community
ISBN: 81-208-1309-X
Rs. 245

Page #3

HARMLESS SOULS

Page #4

LALA SUNDAR LAL JAIN
RESEARCH SERIES
General Editor
PROF. DAYANAND BHARGAVA
Editorial Board
PANDIT DALSUKH BHAI MALVANIA PROF. KIYAOKI OKUDA DR. NATHMAL TATIA DR. PADMANABH S. JAINI PROF. T.G. KALGHATGI PROF. COLETTE CAILLAT
PROF. ERNEST BENDER SHRI K.C. LALWANI PROF. KLAUS BRUHN SHRI N.P. JAIN SHRI SHANTI LAL JAIN
VOLUME IX

Page #5

| HARMLESS SOULS
Karmic Bondage and Religious Change in Early Jainism with Special Reference | to Umāsvāti and Kundakunda
W.J. JOHNSON
MOTILAL BAÑARSIDASS PUBLISHERS PRIVATE LIMITED • DELHI

Page #6

First Edition: Delhi, 1995
© MOTILAL BANARSIDASS PUBLISHERS PRIVATE LIMITED All Rights Reserved
ISBN: 81-208-1309-X
Available at: MOTILAL BANARSIDASS 41 U.A. Bungalow Road, Jawahar Nagar, Delhi 110 007 120 Royapettah High Road, Mylapore, Madras 600 004 16 St. Mark's Road, Bangalore 560 001 Ashok Rajpath, Patna 800 004 Chowk, Varanasi 221 001
PRINTED IN INDIA BY JAINENDRA PRAKASH JAIN AT SHRI JAINENDRA PRESS, A-45 NARAINA, PHASE I, NEW DELHI 110 028 AND PUBLISHED BY NARENDRA PRAKASH TAIN FORM MOTILAL BANARSIDASS PUBLISHERS PRIVATE LIMITED,
BUNGALOW ROAD, DELHI 110 007

Page #7

Sub
"/
Preface Abbreviations
Introduction
CONTENTS
Part I - EARLY JAINISM H 1
Sub
-1 Bondage and liberation according to the early Svetambara
canon
1.1 Early Jainism - Primary sources and chronology
4
1.2 The force of activity (yoga) in bondage: action and intention 5 1.3 Some early Buddhist and early Jaina attitudes to bondage
compared
1.4 Himsä and the ascetic
1.5 The status of the householder according to the earliest texts The status of the householder Bo
i)
ii) Merit and rebirth
1.6
Parigraha and the origins of the kaṣāya doctrine i) Parigraha 301
**
ii) Kaṣāya
iii) Kaşaya and parigraha
1.7 Activity and karma before Umāsvātį.
i) The meaning of arambha Box
ii) Ascetic and non-ascetic actions and karma
Part II - UMĀSVĀTI'S JAINISM HI

ix
xi

1
120
11
2223
26
31
34
37
33838
2 The mechanism of bondage according to the Tattvārtha Sūtra 2.1 Umāsvāti's Jainism - Primary sources and chronology
46
2.2 The mechanism of bondage according to the Tattvärtha Sutra 47 2.3 Sāmparāyika and īryāpatha karma in the Tattvärtha Sutra 2.4 'Activity' in the Tattvärtha Sūtra
57
64
2.5 Parigraha: the householder and the kaṣāya doctrine
72
40

Page #8

vi Harmless Souls
3 Conclusions
Part III - KUNDAKUNDA: THE PRAVACANASĀRA N. I
4 Kundakunda: content and context
4.1 Kundakunda - Primary sources and chronology
4.2 Upayoga
i) Upayoga before Kundakunda
ii) Upayoga according to the Pravacanasara
(i)
ו.
5 The mechanism of bondage according to the Pravacanasāra 5.1 The mechanism of bondage
5.2 Moha
5.3 Himsă in the Pravacanasāra
The role of himsa in the Pravacanasāra
ii) Himsā, moha and upadhi
iii) Himsā and compassion
i) Dhyāna
ii) Dhyana and jñāna
iii) Jñāna
6.5 The rationale for external ascetic practice according to the
Pravacanasara
6.6 Socio-religious roles in the Pravacanasara
Part IV - KUNDAKUNDA: THE SAMAYASĀRA
7 Kundakunda: definitions and truths
7.1 Samayasara
i) Introduction
ii) Samaya
6 The mechanism of liberation according to the Pravacanasāra
6.1 Căritra
6.2 Sāmāyika
6.3 Tapas and dhyāna
6.4 Meditation in the Pravacanasāra
7.2 Vyavahāra-niscaya: the two truths doctrine
7.3 Two 'two truths' doctrines
79

91
97
97
111

124
149
156
168
176
185
188
196
201
205
210
217
224
H2
231
233
239
254

Page #9

Contents
vii
7.4 Dvikriyāvāda
8
The mechanism of bondage according to the Samayasāra i) Bhāva ii) The role of intention (adhyavasāna) in bondage
267
272
9 The mechanism of liberation according to the Samayasāra 9.1 Liberation in the Samayasāra i) Knowledge and the knower
275 ii) Self-realisation
278 iii) Discriminative knowledge
283 iv) Renunciation of bhāva
288 19.2 Linga, practice and the path to liberation
295 19.3 Liberation according to Kundakunda: some conclusions 303
"Table
Appendix 1: Niyamasāra Appendix 2: Pañcāstikāya Appendix 3: Pravacanasāra Appendix 4: Samayasāra Bibliography Glossary and Index
311 312 313 314
321
332 343

Page #10

Page #11

Preface
This book is a revised version of my thesis, 'The Problem of Bondage in Selected Early Jaina Texts', approved for the D.Phil. degree at Oxford University in 1990.
I am grateful to Wolfson College, Oxford, where the original thesis was written, for electing me Michael Coulson Research Fellow in Indology from 1991-1992, thus enabling me to begin work on the revised version. I am also most grateful for the support of the Boden Fund, without which I would have been unable to complete the original thesis, and which also made a grant towards the cost of producing the typescript. The Bhogilal Leherchand Institute of Indology, Delhi, provided me with a research scholarship to study in their library during the early stages of revision in 1990. I take this opportunity to thank Mr. Narendra Prakash Jain of Motilal Banarsidass, Mr. Raj Kumar Jain, and the administrators and staff of the Institute for their hospitality.
My intellectual and academic debts are many - to Alan Williams and Partha Mitter, who encouraged me as an undergraduate at Sussex to believe I could go on to do research, to Jim Benson who taught me elementary Sanskrit so intensively, and to Alexis Sanderson who taught me more Sanskrit and gave generous advice on many subjects. I should also like to thank the examiners of the thesis, Friedhelm Hardy and Roy Norman for their criticisms and suggestions, which I have attempted to incorporate into the revised version. My principal and overriding debt is to Richard Gombrich, who taught me Prākrit and gave me a level of encouragement, advice and support far in excess of the most rigorous standards of professorial responsibility. I could not have hoped for a better supervisor or to have worked under a more inspiring scholar.
My greatest personal debt is to my wife, Patricia, who has more than once suspended her own academic work to

.
.

Page #12

x
Preface
enable me to press on with mine. To her, and to our son, Jonathan, I dedicate this book.
Cardiff 1994

Page #13

Abbreviations
[material in square brackets refers to entries in the Bibliography]
AN Ātmakh
CPD Das
DN JGM
JPP
MN Niy PTS Pañc Pravac Sam
Anguttara Nikāya Ātmakhyāti (see Kundakunda (4), Samayaprābhrtam] Ayāramga Sutta (unless otherwise stated, references employ Jacobi's numbering system, duplicated by Schubring and Bollée) A Critical Pāli Dictionary Dasaveyāliya Sutta (references are to Leumann's ed., unless otherwise stated) Digha Nikāya Jaina Grantha Mālā ed. of Samayasāra (see Kundakunda (4), Samayaprābhrtam] The Jaina Path of Purification (see Jaini, P.S. (1979)] Majjhima Nikaya Niyamasāra (see Kundakunda (1)] Pali Text Society Pañcāstikāya [see Kundakunda (2)] Pravacanasāra (see Kundakunda (3)] Samayasāra
[see Kundakunda (4)] (references are to Chakravarti's ed., unless otherwise stated) Sacred Books of the East Sacred Books of the Jainas Samyutta Nikāya Sarvārthasiddhi (see Pūjyapāda] Sūyagadamga Sutta (unless otherwise stated, references employ Jacobi's numbering system, duplicated by Schubring and Bollée) Tattvadipikā
(see Kundakunda (3)] Tattvārtha Sūtra (numbers in brackets indicate the Śvetāmbara text, otherwise reference is always to the Digambara version, as employed in the Sarvārthasiddhi)
SBE SBJ SN SS Sūy
TD TS

Page #14

xii
Abbreviations
TV
Utt
Tātparya-vrtti (see Kundakunda (4), Samayaprābhrtam] Uttarajjhayana Sutta (references are to Charpentier's ed., unless otherwise stated) Vinaya Viyāhapannatti (numbering is that employed by Deleu [1970]) Yoga Sūtras (see Patañjali]
Vin

Page #15

PARTI EARLY JAINISM
Introduction
The ascetic practices of early Jainism are conditioned by three different but intertwined concepts or beliefs. First, that virtually all matter is alive, in the sense of containing life-monads or souls. Second, that doing harm to living beings is wrong. Third, that actions inevitably have results which affect the future condition and future births of the actor.
If souls (jīva) are ubiquitous, then it is clearly very difficult to do any action at all without harming them. Such harming action (himsā) is believed to result in karmic bondage; that is to say, the soul is invaded and weighted down by subtle matter which ensures that at death the jiva is reborn in this or another world (samsāra), rather than rising to a state of liberation and omniscience at the top of the universe. Consequently, the more harm one does, the heavier the bondage and the worse the rebirth. In short, according to these beliefs, an ordinarily active life in the world will almost inevitably involve too much hiņsā, and therefore bondage, for the actor to have any realistic hope of even a good rebirth, let alone liberation.
On the other hand, to avoid such bondage, it is essential to observe the vow of non-injury (ahimsā) towards all creatures. The central concern of Jaina practice, therefore, is to establish a means of conducting oneself which (ideally) entails no himsā and thus no further bondage. (An important secondary concern is, of course, to get rid of the karma one has already accumulated.) Given the above conditions, this is clearly a very difficult undertaking, requiring special, ascetic restraints. It is particularly problematical for ordinary householders; in fact, prima facie, 'lay Jainism' would seem to be a contradiction in terms.

Page #16

2 Harmless Souls
But Jainism did develop as a religion, as opposed to a personal soteriology - a religion which acquired lay followers and then (numerically, at least) came to be dominated by them. Therefore, it is my purpose in the first part of this work to consider the manner in which this religion developed, and I shall do so by examining both the ways in which the needs and circumstances of the laity were reconciled (in so far as they were) with early, purely ascetic doctrines, and the further problems to which such an enterprise inevitably gave rise.
The first textual synthesis of Jaina doctrine, Umāsvāti's Tattvārtha Sūtra, attempts just such a reconciliation of ascetic and lay concerns. It does so, as we shall see, through a mixture of doctrinal reformulation, doctrinal rejuxtaposition, and doctrinal expansion. Crucial to the new synthesis is the postulation and development of a proper (i.e. technical) doctrine of the mechanism of bondage. That is to say, the way in which karmic matter is attracted and bound to the soul is precisely delineated for the first time. This, however, gives rise to some internal contradictions: the new doctrine is apparently incompatible with certain aspects of canonical teaching. But it is precisely through the examination of these contradictions that it becomes possible to infer what is significantly new about the Tattvārtha Sūtra's mechanism of bondage. (For the content of that canonical teaching, and the ascetic practices which are founded on it, I shall refer to the earliest parts of the Svetāmbara canon, contrasting the doctrines found there with their reformulation and transformation in the Tattvārtha Sūtra.)
In short, by examining the question of what is perceived to be the immediate cause of bondage, and considering how the answer changes throughout Jainism's early history, it is possible to chart the way in which the religion grew beyond the extreme asceticism of its roots and delineate some of the incompletely resolved tensions to which that growth gave rise. The ways in which apparently insuperable theoretical contradictions are overcome, or evaded, in the

Page #17

Introduction 3 actual practice of ascetics and laity, will be discussed in the conclusions to Part I.
I begin, however, with an account of the earliest detectable Jaina view of the causes of bondage, as found in the most ancient parts of the Svetāmbara canon.

Page #18

Bondage and liberation according to the
early Śvetāmbara canon
1.1 Early Jainism - Primary sources and chronology
To determine the earliest Jaina view of the mechanism of bondage, I have referred to the Āyāramga, Sūyagadamga, Dasaveyāliya, and Uttarajjhayaņa Suttas, although some reference is also made to later canonical texts, especially to the fifth Anga of the Svetāmbara canon, the Viyāhapannatti (Bhagavai).1
The Jaina canon presents considerable problems of chronology; of the texts named above, however, the Ayāramga and the Süyagadamga are almost universally agreed to be the oldest on grounds of language and metre. And within these two, the first suyakka mdha (śrutaskandha) of each text is thought to contain the earliest material in each case. There is less agreement about the Dasaveyāliya and Uttarajjhayaņa, although both Schubring and Alsdorf accept them, at least in part, as the oldest texts in the canon, along with the two already named.? Alsdorf remarks that nothing really contradicts the idea that the doctrines contained in these most ancient Jaina texts go back to the time of Mahāvīra, and that even there one does not get to their roots. Any precise dating,
1 When I use the term 'canon' I am, of course, referring to the Svetāmbara texts. The Digambaras, as is well-known, deny that a canon survives. See, for instance, JPP pp. 49-52.
2 See, for example, Schubring 1962, p. 81; JPP p. 53; cf. Alsdorf 1965, p. 28, and in general Alsdorf's analyses of āryā metre to establish chronology: āryās indicating more recent material, sloka or tristubh that which is earlier (e.g. 1966 and 1962-63). On the history of the early canon, see also Alsdorf 1977. . 3 Schubring 1962, p. 81; Alsdorf 1965, p. 28.
4 Alsdorf 1965, p. 28.

Page #19

Early Jainism 5 however, is clearly not possible. It is not even known when the Śvetāmbaras first began to write down their canon. P.S. Jaini suggests some time prior to the second council at Mathura in the fourth century CE. But the final redaction was not made and committed to writing until the end of the fifth or beginning of the sixth century.5
Later canonical texts raise similar problems of chronology, which I shall not consider here. However, further research into these works may well help to lay bare the process of doctrinal change which occurs between the Āyāramga Sutta and Umāsvāti. For my present purposes it is sufficient to show that such a change has taken place, and the longer the period to have elapsed between the canonical material and the Tattvärtha Sutra the more clearly that change is delineated. For that reason I have chosen here to compare the teachings of the Tattvärtha Sūtra with those in what are generally acknowledged to be the earliest extant canonical texts.
1.2 The force of activity (yoga) in bondage: action and intention
K.K. Dixit, characterising Āyāramga 1 and Suyagadamga 1, which are generally admitted to be the earliest surviving Jaina texts, says that they put an unconditional emphasis on world-renunciation, extol the life of the monk, and have 'nothing but condemnation for the life of the householder'.6 Under such conditions it is difficult, if not impossible, for any community of monks 'to forge special links with any community of householders'. The ascetic's life, which is so hard to follow, is designed to reduce monks to a minimum of dependence upon lay society. The texts connect parigraha (attachment to worldly things / 'possession') with arambha ('violence') and treat them as
5
See JPP
6 Dixit 1978, p. 4.
7
Ibid. p. 5.

pp. 51-52; Doshi pp. 26-27.

Page #20

6 Harmless Souls the two most fundamental sins; the former is the proximate cause of sinful activity and the latter the immediate cause. 8
The objects of parigraha may be either animate or inanimate - 'material goods and social relatives'.9 To satisfy the demands of this attachment to worldly things, ārambha is undertaken. The objects of ārambha are the six types of living beings (i.e. the totality of transmigrating jīvas): 1) trasa - 'mobile' jīvas with more then two senses (including humans, animals, birds, insects, etc.), and 2-6) sthāvara - 'static' jīvas with only one sense (i.e. those in earth, water, fire, air, and plants).10
As Dixit remarks, this understanding must have accentuated the strong ascetic tendency of Jaina speculation. 11 That is to say, one cannot undertake activity which manipulates earth, etc., indeed, given the ubiquity of jīvas, almost any activity is liable to be harmful in some way or other. The later idea formulated to accommodate the laity - that violence done to the sthāvara beings is less sinful than that done to trasa beings - is not found in these texts. 12
In the earliest Jaina textual treatment of 'ethical' problems, it is said of the evil-doer that, typically, he either commits a particular evil act, or has it committed by someone else, or approves of it when it is committed by someone else. Thus at Sūyagadamga 1.1.1.3. we read: 'If a man kills living creatures himself, or causes them to be killed by others, or authorises / allows / approves their killing, animosity will increase for himself 13
8 Ibid. A vrabh has the basic meaning of 'to undertake', but in Middle Indo-Aryan dialects it falls together with ā slabh, 'to kill'. So there is an ambiguity in Prākrit and a tendency for ārambha to mean 'killing'. For a further discussion, see pp. 38-39, below.
9 Dixit 1978, pp. 6, 18-19. 10 See Āy. 1.1.2-7. 11 Dixit 1978, p. 6.
12 Ibid.
13 sayam tivāyae pāņe aduvā annehim ghāyae |

Page #21

Early Jainism 7 Dixit speculates that this formulation of the characteristic behaviour of the evil-doer went through a . series of changes in the following manner.14 In the first formulation, the evil-doer either:
1) commits a particular (evil) act, or 2) has it committed by someone else, or 3) approves of (or allows?) an (evil) act when it is
committed by someone else.
In the second formulation, the evil-doer commits an (evil) act either:
1) through body - he does it himself, or 2) through speech - he employs an agent to do it, or 3) through mind - he approves of an (evil) act when it is
committed by another.
The relationship between the two formulations was, according to Dixit, eventually 'forgotten', and it became 'customary to speak of a triple evil act committed in a triple manner'.15 That is to say, an (evil) act committed through:
- by oneself 1) body - by one's agent
- by someone else with one's approval (This corresponds to the second formulation in its entirety.)
- by oneself 2) speech - by one's agent
- by someone else with one's approval
hanamtam vānujāņāi veram vaddhei appaņo || Sūy. 1.1.1.3.|| Bollée's edition, 1977. Tieken 1986, p. 12 ff., offers a different interpretation of this; cf. Bollée's trans., 1977, p. 54. Bollée also gives an alternative translation for veram vaddhei appano - 'his "sin" increases'. Cf. Āy. 1.1.1.5.
14 Dixit 1978, p. 88. 15 Ibid.

Page #22

8 Harmless Souls
- by oneself 3) mind - by one's agent
- by someone else with one's approval.
What is the precise meaning of this? Is an evil act a triple act - does it require all three components to be evil - or is each component evil in itself? Crucially, is component 3), 'mind', a necessary condition for an evil act to take place?
This becomes clearer if we look in more detail at one of the texts cited by Dixit, Dasaveyāliya 4.16 In Prākrit this reads:
iccesim chamham jiva-nikāyāņam neva sayam daņdam samārambhejjā nevannehim daņdam samārambhāvejjā dandam samārambhante vi anne na samaņujāņejjā
The key word here is the ambiguous samaņujānejjā. (Skt. samanujānīyāt - (sam-) anu-vjñā) - 'to fully permit, or allow or consent to, wholly acquiesce in, or approve of. 17 Dixit translates this term with the sense of approval - i.e. one should not approve of a violent action which has been
16 Leumann's ed., bottom of p. 6.
17 Monier-Williams, Sanskrit Dictionary. Cf. Pāli samanuññā - 'approval' - used, for example, at Samyutta Nikāya 1.1 and Majjhima Nikāya 1.159. Also cf. Pāli anujānāti - 'to give permission', 'grant', 'allow' (e.g. Vinaya IV, 225, 25: ekam me itthim anujānātha, 'allow me the power over her'); but when the Buddha speaks in Vinaya, the meaning is near to 'to ordain or prescribe' - see CPD and, for example, Vin. II, 254, 6: na Bhagavā anujānāti mātugāmassa ... pabbajjam. Compare also the formulation of the first pārājika offence in the Suttavibhanga: e.g. 1.10.26:
sādiyi tvam bhikkhū 'ti nāham bhagavā sādiyin ti anāpatti bhikkhu asādiyantassā 'ti 'Monk, did you consent?' 'I did not consent, lord,' he said. 'There is no offence, monk, as you did not consent.'
(Horner's trans. p. 62, etc.)

Page #23

Early Jainism 9 committed by someone else. 18 Schubring, however, translates na samanujānejjā as '(the monk) should not allow others who perform a violent action to do so'. 19 The complete passage reads:
Towards these six groups of souls (i.e. all jīvas ) he should not perform any act of violence himself, nor cause it to be performed by others, nor allow others who perform it to do so.
Schubring's version is clearly the more uncompromising of the two. In his reading a proper act of 'mind' is not simply a matter of disapproval (which is internal, a question of attitude), rather it is a matter of not allowing other people to perform himsā. It is possible that the Jains themselves, or some Jains (the ambiguity of the term permits different interpretations and so different responses to the viotence of others), started with the uncompromising sense - one should not allow others to commit violence if one is aware of their action - and later internalised the idea to a matter of attitude, of approval or disapproval. (Just how a monk could prevent others committing violence without causing violence himself, even if, as is possible, the injunction applies only to preventing his fellow monks from transgressing, is clearly problematic.)
In this respect, the evidence of textual passages relating the Jaina attitude to Brahmanical ritual is interesting. As P.S Jaini points out, Jaina attacks on Vedic sacrifice have at times 'reached the proportions of a crusade'.20 In contrast to incidents in the Buddhist texts, where Brahmans seek out the Buddha to engage him in debate, Jaina stories frequently tell of some kind of active protest against, or interference in, Brahmanical rites. This indicates that the
18 Dixit 1978, pp. 88-89. Cf. Norman p. 14, who, in translating Utt. 8:8, renders na ... anujāne as: 'One should not approve ...'.
19 Leumann's ed., p. 84. 20 JPP p. 169.

Page #24

10 Harmless Souls uncompromising reading of sam-anu-vjñā may have been that of some early Jaina ascetics. For unless the Brahmanical rite was responsible for an influx of karma into the Jaina beholder of the ritual simply through the fact of his witnessing it, why should he be so vehement in his protest? If it were merely the Brahmans' act of violence, causing influx of karmic matter (āsrava) into them (the Brahmans), the Jaina could have let it pass. But it is clear that there is some sense in which the Jaina considers himself personally responsible for it. Belief that an evil act can be committed through the activity of mind explains this. (It may also be the case that the ascetic's attitude is based on compassion for the suffering jiva which is the sacrificial material, although such a reason is not prominently advanced in the texts.)
The passage from Dasaveyāliya 4 continues:
While I live I (shall) not act (violently) in any of three ways, i.e. with mind, speech, and body, nor shall I authorise such action, nor allow another person to act so.21
In contrast to the suggested original meaning of the formula - that acts of body, speech, and mind all referred ultimately to particular physical actions -, this passage introduces a properly mental element. However, this is still a long way from the idea that it is mental attitude alone which is really instrumental in bondage. It is not necessary for all three components, the physical, the vocal, and the mental, to be present for an evil act to be committed; any one of these alone constitutes such an act. Violence is still evil and thus binding, whatever the mental attitude or intention of the actor. Nevertheless, if, as seems likely, the uncompromising reading of sam-anu-vjñā as 'allowance' was eventually relinquished entirely in favour of 'approval',
21 jāvajjīvāe tiviham tivihenam manenam vāyāe käenam na karemi na kāravemi karentam pi annam na samanujānāmi - Leumann's ed. pp. 7-8.

Page #25

Early Jainism 11 then a precedent was established (albeit initially only with regard to the actions of others) in which attitude or intention alone was enough either to cause bondage or, more importantly, to evade it, regardless of what happened physically. A mental event was judged to have soteriological significance regardless of what happened in the external world. Thus, as in the passage cited above, which constitutes the first great vow (Dasaveyāliya 4), it became possible to speak of performing a mental act of violence.
Prima facie, this would seem to increase the difficulties involved in undertaking the vow of ahimsā, since now not only actually accomplished physical acts but also mere 'evil' thoughts are defined as himsā. However, once a mental element is introduced into himsā, the way is open to introduce the further idea that acts are only 'evil' - and so karmically binding - when there is an element of intention: a mental component (viz. 'passion') has to be present. And as will be made clear, this was precisely what happened as Jainism expanded into a religion with a strong lay following. This process will be examined below, along with its doctrinal and social ramifications, but first of all it is necessary to consider in more detail what counts as a harming action according to the earliest Jaina texts.
1.3 Some early Buddhist and early Jaina attitudes to
bondage compared
It is useful to begin by examining the contrasting attitudes of Buddhists and Jains to what is really instrumental in bondage. Such a comparison reveals very clearly what is most important in this respect according to the earliest Jaina canonical texts, namely, physical action. Consequently, it also throws light on the original meaning of 'yoga', which in the Tattvārtha Sūtra is given as the technical term for all forms of physical, vocal and mental

Page #26

12 Harmless Souls
action.22
In the Upalisutta of the Majjhima Nikāya (56), which describes the conversion of Upāli, a lay disciple of Mahāvīra, by the Buddha, a dispute arises over whether the sins of the mind, as the Buddha teaches, or the sins of the body, as the Nigantha Nataputta (Mahāvira) contends, are the heaviest. The Buddha asks Tapassi (a Jain) how many kinds of acts 'effect and start Demerit, according to Nataputta the Nigantha'. Tapassi replies: 'It is not his usage, Gotama, to employ the term "acts"; he speaks of "inflictions" (daṇḍa)' (namely, those of body / deed, word and mind).23
Each of the three kinds of danda is agreed to be distinct from the other two. Mahāvīra is reported by Tapassi to state that, of the three, those of deed (kāyadaṇḍa - i.e. of body) are the heaviest. But the Buddha replies that those of mind are the heaviest; and rather than 'danda', he prefers to use the term 'kamma'. Thus, according to the Majjhima Nikaya, the Jains give a negative gloss to the word for activity itself. (Although, as Jacobi points out, the word kamma occurs in the Jaina sutras too, in the sense of 'deed';
22 See p. 47ff., below.
23 Pāli: kati pana Tapassi Nigaṇṭho Nātaputto kammāni paññāpeti pāpassa kammassa kiriyāya pāpassa kammassa pavattiyā ti... Na kho āvuso Gotama āciņṇam Nigaṇṭhassa Nātaputassa kammam kamman-ti paññāpetum, dandam danḍan ti kho avuso Gotama ācinnam Niganthassa Nataputtassa paññāpetun ti - Majjhimanikāya 1, 372.
The sense of danda here would seem to be 'hurtful physical acts', i.e. 'violence'. Cf. Dhammapada 129:
sabbe tasanti dandassa sabbe bhāyanti maccuno | attānam upamamṁ katvā na haneyya na ghataye ||
'All men tremble at violence, all men fear death. Likening others to oneself, one should neither kill nor cause to kill.'
According to Chalmers (PTS ed. of MN, p. 267), Buddhaghosa 'says that the Jain idea was that citta (the mano-daṇḍa) did not come into bodily acts or into words, which were irresponsible and mechanical, like the stirring and soughing of boughs in the wind'.
S

-

Page #27

Early Jainism 13 danda, however, is at least as frequently used.)24
The term preferred by the Buddhists - kamma - is in essence a more neutral term than danda, in that action is either sinful or not, depending upon intention, i.e. the karmic quality of any action is determined by the quality of volition (cetanā) underlying it. Nothing can be called karmically wholesome (kusala) or unwholesome (akusala) independently of volition; acts in themselves are karmically indeterminate (avyākata).25 Whether a particular volitional state is karmically binding or not depends on the absence or presence of lobha (greed), dosa (hate), and moha (delusion) - perhaps comparable to the 'passions' (kaşāyas) of classical Jaina thought. For this reason, the Buddha can state explicitly that 'volition, O monks, is what I call action (cetanāham bhikkhave kammam vadāmi), for through volition one performs [significant] action by body, speech or mind'.26
From the above, it is clear that for the early Jains physical activity is, by definition, 'hurtful' and thus binding, whereas for the Buddhists it is only binding if accompanied by the mental factors of lobha, dosa and moha. And it is perhaps significant that it is a lay disciple of Mahāvīra's who is converted by the Buddha, since the Buddha's view of what is karmically binding, as represented in the Upālisutta, is clearly more compatible with lay life than the view attributed to Mahāvīra. . It is also interesting to note that 'yoga', the Jaina term
24 Jacobi 1895, intro. p. xvii. See, for instance, Sūy. 2.2 (p. 357ff. in Jacobi's trans.), where thirteen ways of 'committing sins' are treated of, and where the first five are danda-samādāne and the rest are kiriyāthāne (i.e. kriyāsthāna). Cf. also Sthânănga Sūtra (third uddeśaka) where, according to Jacobi (ibid. p. xvii), the doctrine of the three danda is expressed in nearly the same words. Norman p. 15, translating Utt. 8:10, renders damda as 'punishment'.
25 See Buddhist Dictionary, p. 122ff. > 26 Anguttara Nikāya 6.63, quoted in Buddhist Dictionary, p. 92.

Page #28

14
Harmless Souls
for 'activity', is used by Buddhists, in the sense of 'yokes' or 'bonds', as a synonym for the four asavas, the four 'cankers' or 'corruptions' - viz. kāmāsava - the canker of sense desire, bhavāsava - of (desiring eternal) existence, diṭṭhāsava - of (wrong) views, avijjāsava - of ignorance.27 Thus yoga is that which binds for both the Buddhists and the Jains; however, for the former, it is clearly mental events which bind (kāmāsava, etc.), whereas for the latter, it is physical action which is important, for it is physical action which 'causes the influx of karmic matter (āsrava).
Alsdorf has suggested, convincingly, that the use of the term 'āsava' by both Buddhists and Jains, is not a case of one heterodox tradition borrowing from the other, but that when the Buddhists use the term it is a remainder from an ancient, 'primitive' form of a common Indian doctrine concerning the effect and expiation of action - a doctrine which the Jains preserved whereas the Buddhists 'modernized' and 'spiritualized'.28 In other words, the 'original' belief was that the instrument of bondage, of asrava, was physical activity; the Jaina monks retained this idea, whereas the Buddhists redefined yoga and āsrava in terms of mental or 'internal' events.
Etymologically, however, yoga must be the juncture of two things. Caillat defines it as 'the attraction and conjunction' of the material particles which form karma with 'the spiritual monad'.29 P.S. Jaini, referring to the fully developed doctrine of the Tattvärtha Sutra, says that karma generates a vibration (yoga) in the soul which brings
27 See Digha Nikaya 16, quoted in Buddhist Dictionary, p. 27. 28 Alsdorf 1965, p. 4f.
29 Caillat 1974, p. 30; cf. Caillat 1987, p. 511, where she defines yoga as the attraction of subtle matter to the soul through the vibration of its 'soul-points', presumably following the SS on TS 6:1: atmapradeśaparispando yogah. See below, p. 47ff., for my comments on TS 6:1.

Page #29

Early Jainism 15 about the influx (āsrava) of new karmic matter.30 He goes on to say,
The vibrations referred to here actually denote the volitional activities of the individual. Such activities can be manifested through either body, speech, or mind; hence the soul's vibrations are said to be of three types, each corresponding to one of these modalities. 31
As will be made clear, I differ from Jaini on the importance of volition as an element in bondage, particularly in early Jainism; nevertheless, we may readily infer from the above that, in this context, the primary meaning of yoga is the juncture of the soul and matter (which then becomes karma), i.e. bondage. The meaning is then referred back to the cause of that bondage, either vibration of the spacepoints of the soul or the bodily, vocal and mental activities associated with it. However, since activity is the ultimate cause of vibration, it is the activity of the individual which comes to be synonymous with yoga.
For reasons which will become clear, I further suggest that in these earliest Jaina texts the influx of karmic particles and their bondage to the soul is seen as being the inevitable result of activity. In other words, activity is considered binding simply by virtue of its being activity; and the fact that both meanings, 'bondage' and 'activity', can be carried by the same term, 'yoga', bears this out.
In this, I have again followed the Majjhima Nikāya (56) and assumed that for early Jainism volition is not a relevant factor in bondage. But is the account of Mahāvira's attitude to activity given in the Buddhist text really an accurate reflection of the early Jaina position?
In the Süyagadamga (2.6.26-42), the Buddhists are
30 See JPP pp. 105, 112. 31 JPP p. 112.

Page #30

16 Harmless Souls ridiculed for saying (according to the Jains) that it is no sin to cause harm believing that one is not doing so.32 In other words, the Buddhists maintain that mistaking the object of harmful actions relieves the actor of the consequences of his act. Thus killing a man without knowing that he is a man is not sinful; unintentional harm incurs no guilt. For the Jain, however, it is axiomatic that the well-controlled man, one who is careful, does not make such mistakes. He takes care not to harm living creatures whether intentionally or unintentionally. Thus at Sūyagadamga 2.2.3 thirteen kinds of karmas or activities are enumerated, including 'accidental' (akasmāt) sin, and sin committed through an error of sight. And bad karma accrues to people sinning in either of these ways (2.2.11-13).
*Similarly, at Sūyagadamga 2.4.1, Mahāvīra teaches that, 'Even the fool who is unaware of the workings of his mind, speech, and body, and does not see (i.e. register) even a dream, performs evil actions'. 33 This is asserted throughout Sūyagadamga 2.4 against the opponent's (Buddhist's) view that:
If his mind, speech, and body are free from evil, if he does not kill, if he is mindless (i.e. without an internal organ or organ of consciousness), if he is unaware of the workings of his mind, speech, and body, and does not see even a dream, he does not perform evil actions.34
The passage goes on to say that a 'mindless' person still
32 See the passage beginning: piņņāgapiņdimavi... Jacobi's trans. (1895), pp. 414-417.
33 bāle aviyāramaņavayakāyavakke suviņam avi ņa passai pave ya se kamme kajjai - Sūy. 2.4.1
34 asamtaenam maņenam pāvaenam asamtiyāe vaie pāviyāe asamtaenam kāenam pāvaenam ahanamtassa amanakkhassa aviyāramanavayakāyavakkassa suviņam avi apassao pāvakamme no kaijai - Sữy. 2.4.2

Page #31

Early Jainism 17 commits sins of violence, etc., and is thus bound by his actions. Taking the argument a step further, Mahāvīra then states that even ekendriyas, viz. earth-bodies, etc.,
though these beings have neither mind nor speech, yet as they cause pain, grief, damages, harm, and injury, they must be regarded as not abstaining from causing pain, etc. (2.4.9) .... Thus even senseless beings are reckoned instrumental in bringing about slaughter of living beings ... (2.4.10).35
In other words, injury is injury, whatever the motive or lack of motive which accompanies it: what counts is the harmful effect on the object, the injured, not the subjective state of the actor responsible for the injury. Here the ethical, compassionate roots of early Jainism are laid bare: injury is bad in the first place because it is injury to others. It is only with the development of a consistent theory of bondage and liberation that the stress switches from the fact of injury to others to its consequence, namely, selfinjury through bondage. Consequently, the Sūyagadamga does not make it completely clear whether the consequences for the jīva of committing a sin of the kind outlined above are the same whatever embodiment it happens to be in.36
According to the pan-Indian (although not the Buddhist) doctrine of karma and later Jaina theory, karma is only accrued in a human birth; existence in other births is just the reflex of human action. But the Sūyagadamga
35 Jacobi's trans., p. 404, of: jāva no ceva mano ņo ceva vai pāņāņam jāva sattānam dukkhanayāe soyaņayāe jūraņayāe tippanayāe pittaņāyāe paritappaņayāe ... iti khalu se asanniņo vi sattā ... pānāivāe
36 But see Sūy. 2.4.11: 'The venerable One has declared that the cause (of sins) are the six classes of living beings, earth lives etc.' - Jacobi's trans., p. 404 of: tattha khalu bhagavayā chajjīvanikāyāheu pannattā tam jahā pudhavīkāiyā jāva tasakāiyā.

Page #32

18 Harmless Souls appears to be saying that, since intention is irrelevant to sin, and thus to karmic bondage, therefore just as the 'mindless' man binds himself by his actions alone so do jīvas in all other conditions down to ekendriyas (i.e. beings who have no organs through which they can experience passion or intention). This would certainly be consistent with the view that it is action and not intention which is instrumental in bondage, and is a further indication that the instrumentality of passion (kaşāya) is a relatively late addition to Jaina belief.
Thus the ideal Jaina monk, according to Sūyagadamga 2.4.11, is one who abstains from the five cardinal sins and all the vices, who 'does not act or kill'; he is 'wellcontrolled and restrained, avoids and renounces sins, is not active, but careful and thoroughly wise'.37 Comparing this with the standard formula given, for instance, at Sūyagadamga 2.1.50 - 'A monk who does not act, nor kill, who is free from wrath, pride, deceit, greed, who is calm and happy,' etc.38 - the implication would seem to be that wrath, etc. (the four kinds of passion or kaşāya) are not directly instrumental in bondage as such, but that they lead to violent action (or action of any kind?) which is binding. This probably reflects the original connection of kaşāya with bondage. 39
Colette Caillat, in her study of the Cheya Suttas, 40 states that Jaina teachers are anxious to 'redeem and reform even the very impulses and intentions' of the monks. Consequently, they take trouble to 'divine these intentions
37 Jacobi's trans., p. 405 (with minor alterations), of: ...akirie alūsae ... samjayavirayapadihayapaccakkhāyapāvakamme akirie samvude egantapandie ...
38 Jacobi's trans., p. 352, of: se bhikkhū akirie alūsae akohe amāņe amāe alohe uvasamto parinivvude ....
39 See pp. 35-36, below, for a discussion of this. 40 Caillat 1975.

Page #33

Early Jainism 19 correctly and to appreciate exactly the responsibility of the offender'. She concludes that 'One cannot then take the accusations of the Buddhists literally when they accuse the nirgrantha of not according any value to the intentions which motivate the individual'.41 She also says that passages in the Sūyagadamga, such as 2.6.26 and 2.2 (quoted above), 'are not conclusive'. 42
Without entering here into the relative chronology of the Cheya Suttas and their commentaries (from which Caillat draws her evidence) and the earliest portions of the Svetāmbara Canon, I would point out that, according to the early texts, intention is significant in so far as it may lead to or away from physical himsā, but in terms of the mechanism of bondage it is action or restraint from action that counts. For, in any given case, the intention may be good but the fact of physical hiņsā is incontrovertible evidence that it is not good enough. In other words, the emphasis in the earliest texts is not on intention or lack of it as such, but on the degree of direct involvement with himsā. Actions are judged, in the first place, according to their result, not according to the intention of the actor. The latter may be significant before the act, but afterwards it is irrelevant. It is external harm or lack of harm that matters when the soteriological consequences of an action are calculated.
Caillat herself corroborates this in her comments on the Vavahāra Pīthikā, which analyses the acts of the monk into their constituent elements in order to determine to what extent he has sinned.43 As a typical example, she cites the
41 Ibid. p. 108.
42 Ibid. fn. 1, referring to La Vallée Poussin's fn. 3, p. 2 in his edition of L'Abhidharma de Vasubandhu, IV 155, Paris 1923-31, where such passages are cited to support the Abhidharmakośa's interpretation of the Jaina attitude.
43 See Caillat 1975, p. 104ff.

Page #34

20 Harmless Souls case of a monk who picks up, or puts down, a stick. He is totally innocent if in taking hold of it or in laying it down he acts with attention and cleans it. Conversely, he deserves five days of austerity (tapas) if he is guilty of negligence on the first or second point, or on both. Nevertheless, it is still necessary that no creature should have been injured (my emphasis]. Months of tapas have to be observed for harm done to a living creature when one spits, for instance, or if one receives alms in a wet bowl; and the death of a living creature involves complete loss of seniority.44
1.4 Himsā and the ascetic
From the above, it is clear that in the earliest Jaina texts yoga refers primarily to physical action, and that when it is harming it is 'sinful' (i.e. evil action - pāvakamma), and thus binding, regardless of intention or consciousness. For this reason the behaviour of the Jaina monk is characterised above all by physical inactivity and restraint: An early text dealing with ascetic behaviour, the Dasaveyāliya Sutta, makes this very clear.
Dasaveyāliya 4.1 reads: 'He who walks, (stands, sits, and lies down, eats and speaks) carelessly, will hurt living beings. He binds evil karman, that is his bitter reward. '45 And Dasaveyāliya.4.8 states: 'He should walk, stand, sit, and lie down carefully; if he eats and speaks carefully, he does not bind evil karman'. 46
Nothing is said about binding 'good' karman; the important thing is not to bind karman at all (i.e. all karman
44 Ibid. p. 105. 45 Trans. by Schubring, Leumann's ed. 1932, p. 87, of: ajayam caramāņo u pāņa-bhūyāi himsai| bandhai pāvayam kammam tam se hoi kaduyam phalam || 46 jayam care jayam citthe jayam āse, jayam sae jayam bhuñjanto bhāsanto pāvam kammam na bandhai ||

Page #35

Early Jainism 21 is evil in so far as it leads to further bondage). The attitude is that of Ayāramga 1.3.1.4, where killing (chanam) is described as the root of karma (kammamūlam) and rebirth; that is to say, karman is profoundly negative according to these earliest texts.
It is clear that, for the Dasaveyāliya, himsā of any kind results in bondage, regardless of whether such 'harming' is intentional or unintentional; that is to say, there is no such thing as an 'accident' for the Jaina monk, all himsā is due to carelessness. So at Dasaveyāliya 5.1.5 a monk is warned to be careful at all times on his begging tour, since, 'By falling or stumbling a (monk, however) self-controlled, would injure moving or unmoving beings'.47 And at Dasaveyāliya 6.10, it is explicitly stated that, 'As many moving or unmoving beings as there are in the world, (so many of them) a monk should not injure or cause to be injured, either consciously or unconsciously'.48 Thus, among many other restrictions, 'he should duly (and) with exertion inspect (his) alms-bowl and cloth, his bed, the place of excretion, straw or seat' (Dasaveyāliya 8.17), and 'Excrement, urine, mucus, phlegm, (and) filth, he should put away, having, by inspection, found out a pure place' (Dasaveyāliya 8.18).49
Throughout the Dasaveyāliya Sutta, the necessity for the monk to act in a manner which does not cause
47 pavadante va se tattha pakkhalante va samjae | himsejja pāņa-bhūyāim tase aduva thāvare ||
See also Das. 5.1.57-64 for the prohibition on food and drink which might have been mingled with blossoms, seeds, plants, dust and mould, fire, etc.
48 Emphasis added to Schubring's trans. (with alterations) of: jāvanti loe pānā tasā aduva thāvarā | te jāņam ajānam vā na haņe no va ghāyae || 49 Schubring's trans. (with minor alterations) of: dhuvam ca padilehejjā jogasā pāya-kambalam | sejjam uccārabhūmim ca samthāram aduvāsanam ||8.17|| uccāram pāsavaṇam khelam singhāņa jalliyam | phāsuyam padilehittā paritthāvejja samjae ||8.18||

Page #36

22 Harmless Souls
intentional or unintentional himsā is repeatedly stressed (see 4.1-9). This is combined with a constant awareness of the extreme difficulty of treading such a path. At Dasaveyaliya 6.4, it is described as 'the hard, difficult conduct of the Free Ones'.50 And, even more explicitly, in the following verse it is said (perhaps even boasted) that,
In no other (system) has been taught anything which, among worldly people, is (so) hard to carry out; a man may adhere to a great many tenets, (but there has) never been (taught), nor will be (taught to him) anything like our principles. 51
The Дyaramga Sutta is more laconic: 'A very severe religion has been proclaimed'. 52
In all these early monastic rules, nothing is said about intention or attitude as such; for what counts for salvation is physical harm itself; what causes the harming action to arise is of secondary concern. In other words, it is harmful action (himsa) which is directly instrumental in bondage, not attitude. Attitude is only significant in so far as it leads to or away from himsā; it does not cause karmic bondage in itself.
50 Schubring's trans. of: nigganthāṇam durahiṭṭhiyam
51 Schubring's trans. of:
nannattha erisam vuttam jam loe parama-duccaram viula-ṭṭhāṇa-bhāissa na bhūyam na bhavissai || Das. 6.5 ||
The precise meaning of this verse is obscure. Lalwani (1973) gives an even freer translation:
...
'This sort of highly difficult conduct
For the world of human beings
Is delineated nowhere save in nirgrantha philosophy.

āyāragoyaram
For those covetous of mokṣa
Such conduct has nowhere been prescribed in the past Nor will it ever be prescribed in the future.'
The general meaning, however, is not in doubt. 52 Ay. 1.6.4.2: ghore dhamme udīrie.

Page #37

23
Early Jainism 1.5 The householder in the earliest texts
i) The status of the householder Although the doctrine of karma, with the doctrine of rebirth and liberation (mokşa) as its corollary, is the most prominent feature of the earliest Jaina texts, it is not developed systematically and there is no discussion at all of its precise mechanism.53 Moreover, the earliest detectable stream of doctrine holds an uncompromisingly negative view of the householder because he is a householder. The prospect of a better rebirth in heaven or on earth, as a result of good activity which attracts good karma, is hardly admitted, and the four possible births (gatis) seem to have only a theoretical significance at this stage.54 As Dixit puts it, all action leads inevitably to a 'more or less inauspicious' rebirth, and is ipso facto bad.55 This contrasts with the position found in later doctrinal layers where the pious householder and the good monk who is not yet good enough to attain mokşa are promised auspicious rebirths. 56
Given that the ultimate soteriological goal of Jainism is total liberation from samsāra, the idea that any rebirth is relatively undesirable remains a constant component of doctrine. However, what is largely absent from the earliest texts is the idea that there is any gradation or progression through a series of births to ultimate liberation. Instead, what is emphasised is the critical nature of the present birth and, necessarily (since these texts are addressed to ascetics), those kinds of ascetic restraint which will ensure that there is no further rebirth. Thus Āyāramga Sutta 1.6.2, for instance, apparently considers that there are only two possibilities after death: 1) birth among hellish beings
53 See Dixit 1978, p. 9.
54 References to the gatis in the very earliest parts of the canon are few and far between and cannot be dated with any certainty. See, for example, Sūy. 1.2.3.13, 2.2.60ff. and Utt. 5.19ff.
55 Dixit 1978, p. 9. 56 See ibid.

Page #38

24 Harmless Souls and animals, and 2) moksa. The latter will be the condition of the jīva of the ideal monk, and the former that of the jīvas of everyone else, whether householder or monk.57 Āyāramga 1.6.2.1 reads:
Though some know the misery of the world, have relinquished their former connections, have given up ease, live in chastity, and, whether monk or layman, thoroughly understand the law, they are not able (to persevere in a religious life). The ill-disposed, giving up the robe, alms-bowl, blanket, and broom, do not bear the continuous hardships that are difficult to bear. He who prefers pleasures will, now or after a short time, be deprived (of a human body, not to recover it) for an infinite space of time. And thus they do not cross.(samsāra), for the sake of these pleasures which entail evil consequences and are associated with others of their kind.58
It is clear from this that anything short of full mendicancy will entail a long series of miserable, nonhuman rebirths. Similarly, as Dixit points out, 59 at Āyāramga 1.3.4.4, in a sequence which begins with anger (koha) and ends with pain (dukkha), re-birth is described not only as entry into a womb (gabbha), a new birth (jamma), and a new death (māra), but also as characterised by a birth among hellish beings (naraya), animal existence (tiriya) and pain (dukkha).
In this respect, it is significant that the Dasaveyāliya Sutta, for instance, states that for a monk to return to the life of a householder apparently necessarily entails (among
57 See ibid. p. 16. - 58 Jacobi's trans., with minor alterations, of:
āuram logam āyāe caittā puvva-samjogam hiccā uvasamam vasittā bambhaceramsi vasu vā aņuvasu vā jāņittu dhammam ahā-tahā ah' ege tam accāi kusilā vattham padiggaham kambalam pāyapuñchaņam viosijjā anupuvveņa anahiyāsemāņā parīsahe durahiyāsae. kāme mamāyamāṇassa iyāṇim vā muhutte vā aparimāņāe bheo, evam se aạtarāiehim kāmehim ākevaliehim; avinnā c'ee - Āy. 1.6.2.1 - Schubring's text (1910).
59 Dixit 1978, p. 16.

Page #39

Early Jainism 25 other disadvantages) a worse re-birth after death. Dasaveyāliya 11.7 reads: '(To return) means going down (after death)': 60 Such a fate could hardly be construed as some kind of punishment simply because he has given up being a monk, because, since influx of karmic matter (āsrava) is tied to himsā, there is no karmic result from such a change of status in itself: rather, rebirth in hell is here portrayed as the inevitable result of leading the life of a householder.61 For, as Dixit points out, the idea that acts involving the employment of human or animal labour were particularly 'sinful' (i.e. binding) and others less so, does not appear in the earliest texts.62 That is to say, all acts which harmed any of the six types of living beings, whether trasa (mobile) or sthāvara (immobile), were considered to be equally binding. (To be clear about this, the ethical attitude is not so much that a man is as worthless as a mango or a louse, but that a mango or a louse is as important as a man, all jīvas having equal value.) Such a distinction between the binding effects of doing harm to trasa beings on the one hand, and sthāvara beings on the other, only emerged in later Jaina speculation, where the concept of the 'pious householder' is defined as 'one who abstains from all violence done to the trasa beings', that done to sthāvara ones being tolerated.63 As P.S. Jaini puts it, the vow of ahiņsā in its partial, i.e. lay, form applies
60 Schubring translates aharagai-vāsovasampayā as '(To return) means to reach a (place in hell (after death)'. Cf. Āy. 1.6.4.1: 'When they (the disciples) feel the hardships (of a religious life) they slide back, for their love of life. Their leaving the world is a bad leavingJacobi's trans. of putthā v'ege șiyattanti jīviyass' eva kāraņā. nikkhantam pi tesim dunnikkhantam bhavai.
61 Cf. Sūy. 1.3ff. on the potentially fatal temptation to return to lay life. Das. 11 also stresses the social disadvantages of being an exmonk: see, for example, 11.6. See also Olivelle 1974, p. 20, on the general revulsion felt by Indians for the parivrājaka who attempts to return to a society where there is no place left for him.
62 Dixit 1978, p. 6. 63 Dixit 1978, p. 6.

Page #40

26 Harmless Souls 'only to beings with two or more senses (trasa), it is extended for the monk or nun to include the infinitely larger group of single-sense beings (ekendriya) and element bodies (sthāvara )'.64 In later Jainism, it is this complete vow concerning ahimsā which marks the real distinction between the advanced lay-person and the mendicant. In the earliest texts it had similarly distinguished the practice of the Jaina monks from that of all others, but there the distinction was absolute.
Thus, in the early texts, there are passages such as Āyāramga 1.1.2.2-4:
Take note - there are innumerable tiny beings individually embodied in earth. Take note - there are some men who truly control themselves, safeguarding even these beings, while others, (such as the monks of other sects) fail to do so and thus are only pretending to be renunciants.65
In such a context the life of the householder inevitably entails himsā, bondage, and a bad rebirth.
ii) Merit and Rebirth As we have seen, the very earliest texts (i.e. the earliest parts of the Āyāramga, Sūyagadamga and Dasaveyāliya Suttas) are dominated by the idea that virtually all action is harming and therefore binding, since harming is the root of karma.66 Typical in this respect is Āyāramga 1.3.1.3, which asserts that 'action results in misery' (ārambhajam dukkham). In other words, the possibility of some kind of meritorious activity scarcely arises; it is only by restraint from action that one can hope to improve one's condition
64 JPP p. 241.
65 P.S. Jaini's trans. (JPP pp. 241-242) of: santi pāņā pudho siyā, lajjamāņā pudho pasa; anagārā 'motti ege pavayamāņā jam inam viruvarūvehim sattehim pudhavikammasamārambhenam ... anegarūve pāne vihimsai.
66 Āy. 1.3.1.4 - kammamulam ... chaņam.

Page #41

Early Jainism 27 (i.e. one is working towards a smaller quantity of bad karma). Such continuous restraint, however, is only possible at the ascetic extreme inhabited by the Jaina renouncer; for the householder it is, by definition, impossible.67 It is not surprising, therefore, that references to merit (punya) or to activity which leads to a better rebirth in heaven or on earth, as opposed to rebirth in hell or as an animal, are hard to find in this earliest stream of doctrine, and that when they do occur they have an unemphatic and adventitious quality. And it is clear that although, according to some strands of thought, there may be a theoretical possibility of a relatively better rebirth (and perhaps in some cases a practical one for ascetics),68 for householders there is no real possibility of accumulating anything other than more bad karma and the promise of a bad rebirth.
In this connection it is important to remember that what is being reported here is the ascetic's view - the view of the texts. We have no way of telling what views householders may have held. But since the specific category of 'Jaina householder' would seem to be an anachronism if applied to the period in question - for the very reasons being outlined here - then that consideration plays no part in a reconstruction of the earliest layers of Jaina doctrine. This is borne out by passages such as Sūyagadamga 1.7.24-27, where the connection between the giving and receiving of alms and the giving and receiving of the teaching seems to be minimal. Least of all should the monk actually teach the Jain Dhamma, or promise to do so, in order to obtain
67 See parigraha section, p. 31ff., below. 68 See, for instance, Das. 3:14: dukkarāim karettānam dussahāim sahettu ya ke ettha devalogesu kei sijjhanti nirayā ||
'Having done that which is difficult to do, having forborne what is difficult to forbear, some of them (pass) to the worlds of the gods, others attain perfection unburdened [by Karman].' - Schubring's trans. Cf. Sūy. 2.2.74, for example.

Page #42

28 Harmless Souls alms.69 This is interesting because it implies that there was some expectation on the part of householders that alms should be repaid with teaching, combined with the Jaina ascetic's view that any prolonged contact with householders (such as might take place during a sermon) was fraught with potential danger for him. It seems likely, in these circumstances, that the first Jaina ascetics made use of the general (i.e. pan-Indian) cultural perception that it was 'the duty of the householder to feed anyone who came to his door', 70 without subscribing to the additional idea that such giving was a means by which the householder acquired merit. Indeed, as we have seen, the earliest detectable Jaina doctrine of karma leaves no room at all for the idea of meritorious action.
In this connection, there is an interesting verse at the end of a section in the Dasaveyāliya Sutta where the two ideas - the pan-Indian one that there is merit in giving, and the early Jaina one that there is none - are apparently brought together in some kind of compromise. (It should be noted that this verse, Dasaveyāliya 5.1.100, is the last one in a long chapter and could well be a later addition.) This states that, rare as it is to find such people, he who gives for nothing (i.e. the householder) and he who lives for nothing (the monk) both have a good rebirth.71 This
69 See Sūy. 1.7.24: 'He who visits houses where he gets nice food, who professes the Law, desirous only of filling his belly, and brags (of himself) for the sake of food, is not equal to the hundredth part of an Arya.' - Jacobi's trans, of:
kulāim je dhāvai sāugaim āghāi dhammam ayarāņugiddhe ahāhu se āyariyāņa sayamse je lāvaejjā asanassa heū ||.
Cf. 1.7.26: 'The servile man says pleasing things for the sake of food, drink, and other things: but wrong belief and bad conduct are worthless like chaff.' - Jacobi's trans. of:
annassa pāṇassihaloiyassa anuppiyam bhāsai sevamāne pāsatthayam ceva kusilayam ca nissārae hoi jahā pulāe ||
70 Gombrich 1988, p. 75; and see ibid. for reconstruction of contact between early Buddhist monks and householders.
71 dullahā u muhā-dāi, muhā-jīvi vi dullahā || muhā-dāi muhā-jīvi do vi gacchanti soggaim || Das. 5.1.100||

Page #43

Early Jainism 29 seems to mean that householders will only benefit from giving alms if they have no intention of benefiting from them: only alms given with the understanding that the gift is not merit-making are merit-making.72 There is also an ambiguous verse in the Dasaveyāliya Sutta which may suggest that the ascetic should not even accept alms which he knows to have been given for the purpose of obtaining merit (perhaps itself a reflection of the monks' awareness that it is possible to make use of lay misconceptions about the result of giving).73
None of this, however, need seem so paradoxical if it is remembered that the Dasaveyāliya Sutta is a text composed by and for ascetics, and reflects their wariness of any kind of social intercourse. Thus Dasaveyāliya 5.1.100 (see above), rather than teaching a doctrine of motive or intention, simply reflects the monks' concern lest they themselves may come to believe that they are providing, directly or indirectly, something in return for alms, and that they might thereby encourage householders actively to seek them out, or follow them, in an attempt to obtain merit. Such behaviour would inevitably bring them into closer contact with lay life, with its concomitant temptations and himsā-fraught relationships; when possible, it was far safer, therefore, to accept only from those expecting nothing in return. This attitude, enforced by rules of monastic discipline, ensured that it was difficult, if not impossible, for any community of monks, as Dixit remarks, 'to forge special links with any community of householders'. 74
72 Cf. Anguttara Nikāya IV, 60-3, where it is said that the lowest motive for giving alms is with thought of reward in the next life. Quoted by Gombrich 1971, p. 252.
73 See Das. 5.1.49: asanam pānagam vā vi khāimam sāimam tahā|| jam jāņejja sunejjā vā punnatthā pagaim imam ||
Schubring translates this, referring to other verses, as: 'When a monk knows or is informed that food (of any kind) was prepared for the purpose of acquiring merit, he should refuse it', etc.
74 Dixit 1978, p. 5.

Page #44

30 Harmless Souls
Indeed, it is clear that the monks went out of their way to avoid the conditions which might lead to such a relationship.
To summarise, we may suppose that meritorious action and a better rebirth on earth or in heaven as a result of it were concepts familiar to the householders with whom the early Jaina ascetics had their minimal contact; such ideas were part of the general cultural furniture. And although the earliest Jaina doctrine apparently denies the possibility of these consolations to the householder, nevertheless, Jaina mendicants had to take that general cultural view into account when regulating their own relations with lay persons. While their ties with the laity remained so loose, however, it was not necessary for them to make any systematic doctrinal concessions to that view. For the real possibility of a better rebirth for an ordinary lay person to be theoretically established, some doctrine of intention or motive as being, at some level, more karmically significant than action alone would have been required. And as we have seen, in the earliest texts motive is regarded either as totally irrelevant or only important in so far as it helps to engender or inhibit action. (Ideas such as the Buddhist one that there can be karmically wholesome or recommended action [punya-karma] are, of course, entirely absent, as is the related idea that individual monks and the Sangha in general are the 'unsurpassable field of merit' for the laity.)75
Only in Suyagaḍamga 2 (intermittently) and the Uttarajjhayana are there enough references to the selfrestrained householder who achieves rebirth in heaven or as a human to give the impression that this possibility is becoming doctrinally established.76 This is not to claim that there are no references whatsoever to merit and a better rebirth in the very earliest texts; when they do occur,
75See Buddhist Dictionary, 'puñña', p. 80; cf. Collins p. 219.
76 See, for example, Suy. 2.1.13, 2.1.17, 2.2.60ff., 2.2.74, 2.2.78, 2.5.16, 2.7.36, and Ut.. 5.19ff., 7.20ff., 10.15, 12.12-17, 21.94, 36.5053.

Page #45

Early Jainism 31 however, they are isolated and, for the most part, ambiguous.77 Moreover, to talk of a 'pious' or 'meritorious' act remains largely a contradiction in terms, since '
merit', in this context, is usually taken to mean a relative lack of (bad) karma - the result of abstaining from action. 78
The references to merit and the development of the idea in early Jainism would make an interesting independent study, but there is no need here to consider the relative dating of particular sections of any one text to make my point, since the overall tenor of these works (Āyāramga 1, Sūyagadamga 1, Dasaveyāliya) is self-evidently and uncompromisingly anti-householder, as I have demonstrated. That is to say, only from the secondary layers of early material is it possible to adduce firm evidence for particular lay followers who associate themselves specifically with Jaina ascetics (as opposed to all kinds of śramaņas) in the expectation of gaining merit. But even here, the doctrine of merit still lacks a clear, specifically Jaina rationale.
1.6 Parigraha and the origins of the kaṣāya doctrine
i) Parigraha As we have already noted, 79 Dixit has drawn attention to the fact that in the earliest canonical texts the two worst sins are ārambha (violent activity) and parigraha (possession).80 That is to say, they perform the same function as that performed by kasāya (passion), and by the major vows (mahāvratas), in later texts - they 'provide a frame work for the fundamental classification of moral virtues and vices'.81 Moreover, the two are intimately
77 See, for instance, Sūy. 1.2.3.13, 1.7.16, 1.11.17-21.
78 See, for instance, Āy. 1.4.22, Suy. 1.2.13, 1.8.1-2, 1.15.10, and even the relatively late Utt. 29.37.
79 p. 5, above. · 80 Dixit 1978, p. 5.
81 Ibid. p. 7.

Page #46

32 Harmless Souls connected, for parigraha includes that complex of sins which involve a positive attitude towards a particular object, while ārambha comprises those sins which involve a negative attitude. 82 Thus parigraha may be translated as 'attachment to worldly things' or 'possessiveness', as well as 'possession' itself. And such an attitude directed towards a particular object or person necessarily entails an attitude of ārambha towards other objects or persons. Thus all ārambha may be traced back to some case of parigraha . Or as Dixit puts it,
.
Taken as a whole parigraha signifies attachment for things worldly - where things include both the material goods and the social relatives, and ārambha the acts injurious to others undertaken with a view to satisfying the demands of this attachment.83
(Like kaṣāya in its earliest sense (see below], parigraha leads to violence and thus to bondage, but it is not the direct cause of the latter - i.e. it is the cause of violence but not the direct cause of bondage.) Parigraha is thus clearly associated with the life of the householder, for it is the householder who is attached to worldly things. It is a life, moreover, which is inextricably bound up with violence, actual and potential. Thus the Sūyagadamga characterises householders as sārambhā and sapariggahā- 'killers' and 'acquirers of property' in Jacobi's translation.84
82 Ibid. p. 5. 83 Ibid. pp. 18-19. 84 Jacobi 1895, p. 350. See Sūy. 2.1.43-46. Sūy. 2.1.43 reads:
iha khalu 'gāratthā sārambhā sapariggahā, samtegaiyā samaņā māhanā vi sārambhā sapariggahā, je ime tasā thāvarā pāņā te sayam samārambhanti anneņa vi samārambhāvemti annam pi samārambhatam samanujānanti
'Here, indeed, householders are killers (of beings) and acquirers of property, and so are even some Śramanas and Brāhmaṇas. They themselves kill moving and unmoving living beings, have them killed by another person, or consent to another's killing them.' - Jacobi's trans.

Page #47

Early Jainism 33 The Jaina mendicant, on the other hand, is one who 'disowns all things not requisite for religious purposes', or who 'disowns attachment', 85 since 'he who possesses anything sentient or insentient, however slender, or allows / approves of others doing so, will not be delivered from suffering'. 86
In other words, such a person goes on killing and binding new karma.87 Similarly, Sūyagadamga 1.2.2.9-10 states that:
A monk should be completely familiar with the Law, and at the same time no longer be engaged in (worldly) actions (ārambha). (People) who are entangled in thoughts of possessions take care of those possessions; they attain, however, no permanent property. (9)
Know that (property) causes suffering in this world, and that such suffering continues in the next world. Who, knowing this, would be a householder? (10)88
In the Dasaveyāliya Sutta, this is made even more explicit in the statements: 'He who desires to hoard (things) is a householder, not a monk', 89 and 'the Great Sage has
with minor alterations.
85 Alternative translations of pariggaha suggested by Jacobi at Āy. 1.2.5.3.
86 cittamantam a-cittam vā parigijjha kisām avi | annam vā aṇujāņāi evam dukkhā na muccai || Sūy. 1.1.1.2 ||
Bollée's ed., Vol. I 1977, p. 14. Tieken 1986, p. 10ff., offers a very different translation of this; cf. Bollée's trans. (1977), pp. 53-54.
87 See Sūy. 1.1.1.3-5.
88 Based on Bollée's trans. Vol. II 1988, pp. 55-6 of his ed. of Sūy. 1.2.2.9-10:
dhammassa ya pārae muņi ārambhassa ya antae thie soyanti ya nam mamāiņo no labbhanti niyam pariggaham 1911 iha-loga duhāvaham viū para-loge ya duham duhāvaham | viddhamsana-dhammam eva tam ii vijjam ko 'garam āvase ||10|1. Cf. Jacobi's trans. 1895, p. 254.
89 je siyā sannihi-kāme gihi pavvaie na se - Das. 6:19b, Schubring's trans.

Page #48

34 Harmless Souls said that property means delusion'.90 The desire to hoard or store things is to experience greed (lobha),91 and lobha is not simply the worst of the four sins, it in fact includes the other three: 'Anger destroys kindness, pride civility, deceit drives away friends, [but] greed destroys all.'92
It is clear from this that lobha (greed) is synonymous with one meaning of parigraha, and that it is also an inclusive term for the four sins or passions which are later understood as the components of the technical term kaşāya.93
ii) Kaşāya The four 'moral vices' or passions - anger, pride, deceit and greed (krodha, māna, māyā, and lobha) -, which are later called kaşāya (literally, 'stains'), do appear as a set in both the Āyāramga Sutta and the Sūyagadamga, although they are never given any common designation, and there is no extended treatment of them.94
90 mucchā pariggaho vutto ii vuttam mahesiņā - Das. 6:21, Schubring's trans.
91 Das. 6:19a. 92 koho piim paņāsei, māno viņaya-nāsaņo | māyā mittāņi nāsei, lobho savva-viņāsaņo || Das. 8:37||. 93 See below; Cf. JPP p. 177.
94 See Dixit 1978, pp. 7, 15, 19. În Sūy. 1 there are, however, a number of instances where four vices appear under what Dixit calls 'rather peculiar and obscure designations' (Dixit 1978, p. 19). For instance, at 1.1.2.12, the Prākrit terms savvappaga, viukkassa, nūma, and appattiya appear. Jacobi reads these as (Sanskrit) sarvātmaka = lobha), vyutkarşa (=māna), ... = māyā), and appattiya (= krodha?) respectively. Thus his translation reads: 'shaking off greed, pride, deceit and wrath, one becomes free of karman'. (Cf. Bollée, Vol. I (1977), pp. 89-92, on these). Again, at Sūy. 1.1.4.12, a wise monk is enjoined to leave off ukkasa, jalana, nūma, and majjhattha. Jacobi renders these as (Sanskrit) utkarşa (= māna), jvalana (= krodha), ... (= māyā), and madhyastha (= lobha?) respectively. (Cf. Bollée (1977), p. 129.) And at 1.9.11, paliumcana, bhayaņa, thandilla, and ussayana are named as the causes of sin. Jacobi renders these as (Sanskrit) parikumcana (= māyā), bhajana (= lobha), ... (= krodha), and ucchraya (= māna), adding the note that 'these four passions are named here

Page #49

Early Jainism 35 At Sūyagadamga 1.6.26, for instance, it is said of the Arhat that, having expunged the passions which defile the soul, viz. wrath, pride, deceit and greed, he does not commit any wrong, nor does he cause it to be committed. 95 This is typical in that although it sees the four passions as the instigators of wrong (i.e. harmful) behaviour they are not tied directly to bondage in any technical sense, but to himsā; and it is himsā that causes the soul to be bound by karma. This is borne out by passages such as Ayāramga 1.3.4.1, where the conquest of anger, etc. is described as the 'doctrine of the Seer who does not injure living beings and has put an end to acts and to samsāra)': 96 In other words, motivation is still not seen as directly binding in itself; it is the act (killing, etc.) which remains soteriologically crucial. The important thing is control: control of physical action and therefore also control of the passions which may lead to the loss of such physical restraint and the consequent destruction of living creatures. . In the Dasaveyāliya Sutta, the term kaşāya is applied to the four vices (krodha, etc.). Dixit argues that because kaşāya is used in a technical sense here, a relatively late date should be assigned to the passage.97 However, whether Dixit is right about the dating or not - and prima facie his argument seems to be at best circular - kaşāya still has no direct connection with bondage in the sense of providing conditions for karmic matter to stick to the jīva, which is the later technical sense of the term (see below). Thus at Dasaveyāliya 7:57 there is the passage:
from the way in which they are supposed to act upon the soul' (1895, p. 302 fn. 5). These are passages which clearly need further investigation.
95 See Jacobi's trans. (1895, p. 291) of: koham ca mānam ca taheva māyam lobham cauttham ajjhattadosā| . eyāni vantā arahā mahesi na kuvvai pāva na kāravei || Sūy. 1.6.26||
96 Jacobi's trans. (1884, p. 33) of:
se vantā koham ca mānam ca māyam ca lobham ca eyam pāsagassa damsanam uvarayasatthassa paliyantakarassa.
97 Dixit 1978, p. 29.

Page #50

36 Harmless Souls
(He who) speaks after consideration, controls his senses well, has overthrown the four passions, (and) is without (worldly) support, purges (his soul) of the dirt resulting from previous evil deeds (and) may gain this world and the next.
98
And at Dasaveyāliya 8:36-39, the four 'passions' and the ways to subdue them are analysed in more detail as part of a long description of the way in which a monk should behave:
When he wishes that which is good for him, he should get rid of the four faults anger, pride, deceit, and greed which increase evil... Anger and pride, when not suppressed, deceit and greed, when arising: all these four black passions water the roots of rebirth.99
It should be noted that the metaphor here is of passions 'watering' the 'roots of rebirth', not of causing the jiva to become 'sticky'. 100 There is no specific link with karma, and in this case passions (kaṣāya) are again clearly considered to be a contributory factor in bondage and not directly instrumental: the term is evidently not yet being used in its technical sense (i.e. in the sense employed in the Tattvārtha Sutra - see below).
iii) Kaṣāya and parigraha
Richard Gombrich has pointed out the close association in Indian religious thought between asceticism and the idea
98
Based on Schubring's trans. of:
parikkha-bhāsī susamāhiindie caukkasāyāvagae aṇissie |
sa niddhune dhutta-malam pure-kaḍam ārāhae logam iņam tahā param | Das. 7:57 ||.
99
Schubring's trans., with alterations, of:
koham māṇam ca māyam ca lobham ca pāva-vaḍdhanam | vame cattări dose u icchanto hiyam appano || Das. 8:36|| koho ya māņo ya aṇiggahīyā māyā ya lobho ya pavaḍḍhamāṇā | cattāri ee kasiņā kasāyā siñcanti mūlāi puṇabbhavassa || 8:39 ||. Cf. p. 47ff., below.
100

Page #51

Early Jainism 37 that emotion stands as the main obstacle to salvation.101 Jainism is exemplary in that respect. For any system which seeks to regulate physical action to the extent that early Jainism does must also seek to regulate passion or emotion. The two are so closely related as to be virtually interdependent: emotion expresses itself in physical indiscipline, physical indiscipline implies some loss of inner control, whether it be full-blown passion or mere carelessness. Physical and emotional control is therefore paramount. (Again it should be noted that, in terms of karma and avoiding it, it is the physical which takes priority here: one is seeking to control the emotions in order to avoid harmful, and thus karmically binding, actions.)
For the Jaina ascetic, the distinguishing characteristic of the householder's life, with its possessions and the emotions aroused by having and wanting possessions (parigraha and lobha), is lack of control. And as we have seen, according to the Dasaveyāliya Sutta, lobha ('greed') comes to subsume all those emotions which are virtually synonymous with lack of control. This interdependence of 'possessions' (being a householder) and 'possessiveness' (thinking and acting like a householder) are seen as the himsa-causing and thus binding factors par excellence. In this respect, an inadequate monk can all too easily behave like a householder, but a householder cannot (yet) be like a monk. However, the very distinction between 'possessions' and 'possessiveness' does open up at least a theoretical possibilty of non-attachment to possessions, an attitude or intention, having a significant part to play in liberation from karma. And as we shall see, although the interdependence of the two of possessiveness and the householder's life continues to hold good from the ascetics' perspective, for the laity a gap begins to open up. Parigraha thus becomes an all-inclusive term for the inner, emotional reaction to the external world - a reaction
101 Gombrich 1988, pp. 44-45.

-

Page #52

38 Harmless Souls which ensures that one remains bound in samsāra. The beginnings of this internalisation were probably subject to Buddhist influence. For parigraha as 'possessiveness' is very like taṇhā ('thirst', 'craving', or 'grasping'), which for early Buddhism is the karmically significant (i.e. binding) factor. 102 In the Buddhist case, of course, karma is fully internalised to volition; for the Jains it is still ultimately a matter of physical action. Nevertheless, we can see here the beginnings of what later becomes a crucial doctrine for the Jaina lay person. It is only with Umāsvāti, however, as will be made clear, that a technical explanation in terms of the mechanism of bondage is given for the widely held perception that passion is somehow very closely linked to violence and so to continuing bondage.
1.7 Activity and karma before Umāsvāti
i) The meaning of ārambha Monier-Williams gives vrabh / rambh as an early form of the root labh / lambh. A vrabh has the basic meaning of 'to set about', 'begin', 'undertake', whereas, although ā vlabh can also have the meaning 'to commence', it also has the sense of taking hold of (in a physical sense)', and 'to kill', especially 'to sacrifice'.103 In eastern Middle Indo-Aryan dialects rabh and labh fall together (r > 1). Thus there is an ambiguity in Prākrit and Pāli, and for both Jains and Buddhists there is a tendency for the word to mean 'killing'. For the Jains, however, there may have been a doctrinal as well as a linguistic reason for this ambiguity. Ārambha originally meant 'undertaking' / 'beginning', but, given the fact that jīvas were believed to be almost everywhere (vide the sixfold objects of ārambha),104 for the earliest Jaina ascetics virtually any activity was probably perceived as
102 On tanhā, see Rahula pp. 29-34, and Buddhist Dictionary, pp. 218-219.
103 Monier-Williams, A Sanskrit-English Dictionary. 104 See p. 6, above.

Page #53

Early Jainism 39 causing himsā; and thus ārambha as 'undertaking an action' and ārambha as 'killing' or 'violence' were very likely considered synonymous.
This is probably how the term ārambha is to be understood in such passages as Āyāramga 1.3.1.3 (Suttāgame 172):
ārambhajam dukkham inam iti ņaccă, măi pamāi punar ei gabbham, uvehamāṇo saddarūvesu ujjü, mārābhisanki maraņā pamuccai
Knowing the suffering born of action, The deluded and careless person enters a womb again; Indifferent to sounds and forms, upright, Anticipating / fearing Māra one is liberated from death.
Ujjū (Sk. rju) means 'straight' (Sk. varj - 'to stand firm'), and so literally and figuratively 'upright'. Thus, although it can mean 'ethically upright' or 'sincere', aligned here with indifference to sadda (sabda) and rūva (rūpa), ujjū may have the primary meaning of being physically straight or upright (cf. rju-kāya - 'having a straight body'). In other words, it refers to the typical Jaina meditative posture, kāyotsarga - literally, 'abandonment of the body', standing or sitting motionless. 105 If that is the correct reading, then, since the remedy for rebirth as a result of action is total stillness, the implication is that ārambha, in this context, does mean any action. 106
105 See JPP pp. 190, 192, 225.
106 Compare, however, Suttanipāta 8 (Mettasutta) where ujjū (= rju) has the sense of morally upright::
karaniyam atthakusalena yan tam santam padam abhisamecca: sakko ujjú ca sūjū ca suvaco c'assa mudu anatimāni
'Whatever is to be done by one who is skilful in seeking (what is) good, having attained that tranquil state (of Nibbana): - Let him be able and upright and conscientious and of soft speech, gentle, not proud.'

Page #54

40 Harmless Souls
ii) Ascetic and non-ascetic actions and karma 'Arambha', however, soon begins to acquire a more technical meaning, denoting 'purposive', 'deliberate' or 'premeditated action'. This is particularly evident in the fifth Anga of the Svetambara canon, the Bhagavai Viyahapannatti (Bhagavati Vyākhyāprajñapti).107 There, actions (kiriyā, Sk. kriyā) are described as being:
(1) purposive (ārambhiyā kiriyā), (2) appropriative (pariggahiyā kiriyā), (3) emotional (māyā-vattiyā kiriyā), (4) implying nonrenunciation (apaccakkhāṇa-kiriyā) or (5) implying heresy
(micchādamsaṇa-kiriyā).108
In other words, kiriya is here used as a term for all actions, while arambha is reserved for purposive, harming activity directed towards other beings. 109 In spite of the negative nature of the above list (5.6.2), it is clear from other passages in the Viyahapannatti that kiriyā is in itself a karmically neutral term. That is to say, we are now presented with a perception that all action is not necessarily harming, and thus binding, simply because it is action (i.e., the position I inferred to be the earliest ascetic one has now been modified). This is made explicit at Viyāhapannatti 7 7.1, which reads, in Deleu's summary:
When a monk who is closed [against karmic influx] (samvuḍa aṇagāra) moves and handles his equipment in an attentive way (auttam) he commits an action in agreement with his religious duties (iriyavahiyā kiriyā), not a profane action (samparāiyā
Text PTS, eds. D. Andersen & H. Smith, 1913; trans. V. Fausboll, SBE Vol. X, Pt. 2, p. 24 (Oxford 1881).
107 The Viyahapannatti is a long, incoherent compilation of material, some of which is probably very early - see Deleu (1970). 108 Deleu's (1970) summary of Viy. V 6.2; cf. Schubring 1962, p.
199.
109 See Deleu (1970) on Viy. V 7.7. On kiriyā being employed when concrete actions are referred to, see Schubring 1962, para. 99 (p. 198).

Page #55

Early Jainism 41 kiriyā) [see explanation below], because in him the four passions are extinguished (vocchinna) and he acts in agreement with the precepts (ahā-suttam eva rīyai).110
Furthermore, this is not simply a distinction between harming and non-harming activities, it is also an institutional distinction between ascetics and laity: just as īriyāvahiyā kiriyā is the action of a monk conforming to the monastic rules, so samparāiyā kiriyā is the action of a nonascetic or lay person. And while it is considered impossible, by definition, for a non-ascetic or layman to perform an iriyāvahiyā action, 111 it is very easy for an inattentive monk to perform a samparāiyā action. 112 Thus, with monks, actions are said to result either from 'carelessness' (pamāya, Sk. pramāda) or from 'activity' (joga, Sk. yoga). 113 Yoga, therefore, is the minimal action undertaken by a monk following the monastic rule. And, crucially, the karma bound by a monk as the result of this yoga - as a result of 'discharging his religious duties (iriyāvahiyā kiriyā)' - 'is consumed within two samayas' (i.e., virtually instantaneously).114
Deleu remarks that 'Mahāvīra's idea of the iriyāvahiyā action seems to have met with a great deal of incomprehension on the part of his contemporaries', and sometimes even puzzled his own disciples. 115 He goes on to say that the Jaina conception of iriyā-samii / iryā-samiti ('care in walking') was often attacked by rivals, particularly the Ajivikas, although (in Deleu's opinion) 'Mahāvīra's explanation of its real tenor sounds reasonable enough: if a monk hurts some small living being while walking in the prescribed way, the action is still in agreement with his
110 Deleu (1970). Cf. Viy. VII 1.6; I 10.2. 111 See Deleu (1970) on Viy. VII 1.3a. 112 See ibid. on Viy. VII 1.6. 113 See ibid. on Viy. III 3.1c. 114 See ibid. on Viy. III 3.1d. 115 Deleu 1977, p. 190.

Page #56

42 Harmless Souls
religious duties'.116 Furthermore, according to Deleu, such passages as this 'somehow put the old controversial issue regarding the unconsciously committed sin (that divided as is well known the Jainas and the Buddhists) in quite a different light'. 117
There are a number of points to consider here. First, it is not certain that the Viyahapannatti reports the actual words or doctrines of Mahāvira. Even if it does, it is clear that the oldest canonical texts (Āyāramga 1, etc.) contain material which is less sophisticated doctrinally and is probably even older. (If that is the case, it may be that here Mahavira is reforming a previous Jaina or proto-Jaina position, or simply excluding, through unambiguous formulation, what had been a matter for debate in earlier practice.) Moreover, regarding the controversy with the Buddhists, the Jain view is that if a monk is following the discipline properly he cannot, by definition, unconsciously commit a sin. For himsā done while the monk is conforming to irya-samiti is tolerated, not because it is unconscious, but because it is a special case, sanctioned by the monastic rule. The total amount of karma accrued from such actions is agreed to bind for a short enough time to do no real damage to a monk's prospects of liberation, providing he continues to adhere to the discipline. This, of course, allows ascetics a realistic chance of achieving liberation, and may represent the loosening of some previously very tight rule.
The idea that unavoidable injury perpetrated while following the monastic rules, i.e. while acting with rigorous carefulness or awareness, is less karmically significant than injury perpetrated through neglect of the monastic rules, i.e. done out of carelessness, can be found in isolated passages in the early canon. For example, Ayāramga
1.5.4.3 states that:
116 Ibid., referring to Viy. XVIII 8.1. 117 Ibid. Cf. pp. 11-20, above.

Page #57

Early Jainism 43 On occasion, living beings are destroyed by contact with the body of a virtuous monk walking in the prescribed fashion; he will get his punishment in this life. But if it was done contrary to the rules, he should repent of it and do penance for it. Thus he who knows the sacred texts recommends penance combined with carefulness. 118
Again, this is not the same thing as saying that monks are allowed to commit what might be normally understood as 'accidental' injury. The reference is to unavoidable injury done in highly controlled circumstances; this still has karmic consequences, but of a short-term, and thus manageable nature. Although it may be noted that the consequences envisaged in this early text are apparently more serious and longer-lasting than the almost instantaneous entry and exit of karma portrayed in the Viyāhapannatti (see above). 119
The invention of categories of action and karmic 'bondage' which have negligible consequences in effect makes both action and the inflow of karma neutral in themselves. The development of this is perhaps traceable from a passage in the second book of the Sūyagadamga. There, at Sūyagadamga. 2.2.1, thirteen kinds of activity
118 Jacobi's trans. 1884, p. 48, with alterations of:
egayā guṇasamiyassa rīyao kāyasamphāsam (sam)aņucinnā egaiyā pāņā uddāyanti, iha logaveyaņavejjāvadiyam, jam āuttikayam tam pariņņāya vivegam ei, evam se appamāenam vivegam kittai puvvavi* - Āy. 1.5.4.3 (Suttāgame 307). *Schubring's ed. prints veyavi, Suttagame puννανί.
119 Note also Das. 5.1:87-89, where a monk who has collected alms and returned with them to where his guru is, should approach the latter with 'the airyāpathiki formula'. This is Schubring's understanding of iriyāvahiya at Das 5.1:88 (his trans., p. 93, Leumann's ed.). The formula referred to here is probably the third of the āvassava / avasvaka formulae, vandanaga, which is the prescribed way of respectfully addressing a superior upon entering a place (see Schubring 1962, p. 269). Here it is clearly connected with returning from the begging round, i.e. returning from a permitted monastic activity (iryāvahiya). On the 'ritualisation' of monastic discipline, see below.

Page #58

44 Harmless Souls (kiriyā) are named. The last of these is iriyāvahiya (translated by Jacobi as 'actions referring to a religious life'). 120 These 'subtle' or 'insignificant' (suhumā / sūksma) activities of the ascetic, governed by the samiti and gupti, are described in detail at Sūyagadamga 2.2.23.121 Moreover, the influence of the karma attracted to the soul by these actions is said to be only momentary, lasting but three samaya. 122
This passage on iriyāvahiya-kriyā is, however, concluded by the refrain which has accompanied the other twelve kinds of activity, viz. 'Through that something blameable is produced for him' (or as Jacobi translates it, 'Thereby bad karma accrues to him').123 And the inappropriateness of this here suggests that this kind of iriyāvahiya activity may have been a later, somewhat mechanical addition to an original list of twelve bad actions. This conjecture is further borne out by the fact that this passage views the karma acquired from iriyāvahiya actions as being more or less instantaneously destroyed, in contrast to the apparently longer process envisaged by Āyāramga 1 (see above).
Through this kind of development we are brought close to Umāsvāti's distinction between passionate, binding activity and non-passionate, non-binding activity. (The Viyāhapannatti does not, however, contain a 'kaṣāya doctrine' in the technical sense developed in the Tattvārtha Sūtra, although passages like Viyāhapannatti 7 7.1, quoted above, contain all the necessary components.) 124
120 Jacobi 1895, p. 356. 121 Jacobi, in his translation of suhumā kiriyā iriyāvahiyā nāma
(ibid. p. 365), takes iriyāvahiya to refer to subtle actions other than those followed in the discipline, but that seems confused.
122 Cf. Viy. III 1.d, above, and Utt. 29.71 (Jacobi 1895, p.172).
123 vam khalu tassa tappattiyam sāvajjam ti āhijjai - Sūy. 2.2. passim.
124 See also Viy. VIII 8.3a, b, where iriyāvahiya- and samparāiyakamma (as opposed to just -kiriyā) are specifically referred to.

Page #59

Early Jainism 45 The soteriological and social implications of such a distinction are far-reaching, and in order to examine these at greater length I shall now turn to the teachings of the Tattvārtha Sūtra on the subject of long and short-term karma.

Page #60

PART II UMĀSVĀTI'S JAINISM
The mechanism of bondage according to the
Tattvārtha Sūtra
2.1 Umāsvāti's Jainism - Primary sources and chronology
As it represents the first real synthesis of Jaina doctrine, I have, for this part of my argument, used the Tattvārtha Sūtra of Umāsvāti as my point of reference. This is the earliest extant Jaina work in Sanskrit, written between 150 C.E. and 350 C.E.1 Indeed, as P.S. Jaini points out, it 'manages to synthesize virtually the entire Jaina doctrinal system in a mere 350 sūtras'.2 Furthermore, it is not only the one text that both Digambaras and Svetāmbaras recognize as authoritative, but the commentaries on it, whether by Digambara or Svetāmbara authors, 'present almost identical explications of 'Jaina doctrine'.3 (In any case, the differences between Svetāmbara and Digambara doctrine are not significant for the present discussion, which deals with problems fundamental to both traditions.) For that reason I have used the earliest extant Digambara commentary, the Sarvārthasiddhi of Pūjyapāda (Devanandin) (c. fifth century C.E.), in conjunction with the Tattvārtha Sūtra, rather than the alleged autocommentary, the Tattvārthādhigama Bhāsya (also known as the Svopajña Bhāsya). There is in fact
1 See Bronkhorst p.178; also Dixit 1971, pp. 5-8. Ohira pp.135. 137, dates Umāsvāti to the end of the 5th century, but that seems to be too late - see Bronkhorst and Zydenbos, below.
2 JPP p. 82. 3 Bronkhorst p. 178.

Page #61

Umāsvāti's Jainism 47 considerable doubt whether the 'autocommentary' was written by Umāsvāti himself; indeed, Bronkhorst has presented a convincing case for attributing it to a Svetāmbara of the fourth century C.E. (at the earliest) who, in the manner of the Yoga Sūtra and Yoga Bhāsya, incorporated the sūtras into his own work, the Tattvārthādhigama Bhāșya.4 Furthermore, the Sarvārthasiddhi, although composed perhaps a century later than the Tattvārthādhigama Bhāşya, may be using a version of the Tattvārtha Sūtra which is at times closer to the original than that used in the Bhāsya.5 There is also some evidence that the Tattvārtha Sūtra itself was composed in a Digambara milieu, while the Bhāsya has marked Svetāmbara features. In any case, since the commentaries only differ on a few sectarian issues which are not significant for the present discussion, one may take either as authoritative, in the sense that each represents Jaina doctrine in a non-controversial manner which is not at variance with the prima facie meaning of the Tattvārtha Sūtra itself. The Sarvārthasiddhi, however, often develops the logical implications of doctrines which are merely stated or formulated in the Tattvārtha Sūtra, as will become apparent.
2.2 The mechanism of bondage according to the Tattvārtha
Sutra
Tattvārtha Sūtra 6:1-4 (= 6:1-5) states that:
Yoga is the activity of body, speech and mind. [6:1] It is influx. [6:2]
4 Bronkhorst p. 179. See also Zydenbos pp. 9-12, who dates the Tattvārthasūtrabhāsya to the 5th century; for the Sūtra itself, he suggests the end of the 3rd century.
5 Bronkhorst p. 172. 6 Ibid. p. 177; see also R. Williams pp. 2-3. 7 See JPP p. 82.

Page #62

48 Harmless Souls
Good activity is the cause of (the influx of) meritorious (karmic matter), and bad activity is the cause of (the influx of) demeritorious (karmic matter). [6:3 (3/4)]
(There are two kinds of influx, namely) that of persons with passions, which brings about rebirth, and that of persons free from passions which has no effect on rebirth. [6:4 (5)]8
To expand this: Umāsvāti teaches that yoga is the vibration of the soul caused by the activity of speech, the mind and the body. This vibration is in turn the cause of the attraction of groups of karmic particles (although, strictly speaking, this matter only becomes karma as such when it adheres to the soul). Thus, when activity occurs, influx (āsrava) is inevitable: yoga causes the soul to act like a magnet, drawing in karmic matter. All three instigators of yoga can be either good or bad, potentially the causes of the influx of meritorious as well as demeritorious karma.
There are two kinds of asrava, depending upon whether one acts out of passion or not: that activity accompanied by kaṣāya (passions) results in the influx of sāmparāyika karma (rebirth-causing karma), that activity which is free from passions results in the influx of iryapatha karma (short-term karma which has no effect on rebirth). In other words, yoga attracts (karmic) matter to the soul, and kaṣāya causes that matter to adhere to the jiva and to bind it.10
The source materials used in Tattvārtha Sūtra 6 (concerning yoga, äsrava, etc.) are widely dispersed in the canon, deriving from passages in the Bhagavai, the
8
kāyavānmanaḥkarma yogaḥ [TS 6:1]
sa asravaḥ [TS 6:2]
subhaḥ punyasyāśubhaḥ pāpasya [TS 6:3 (3/4)] sakaṣāyākaṣāyayoḥ samparāyikeryapathayoḥ [TS 6:4 (5)] 9 Cf. SS on TS 2:25:
yogo vānmanasakāyavargaṇānimitta ātmapradeśaparispandaḥ 'Yoga is the vibration of the space-points of the soul caused by the group speech, mind and body.'
10
See TS 8:2 (2/3), quoted p. 55, below.

Page #63

Umāsvāti's Jainism 49 Thānamga, and the Uttarajjhayana, among others. 11 These show that at Tattvārtha Sūtra 6:1 Umāsvāti changed the traditional sequence of the threefold yoga (manas, vāc, kāya) into kāya, vāc, and manas. According to Ohira, he probably did this because he attached most importance to kāyikakriyā. 12 But leaving aside the likelihood that the most important element would in fact be placed last, all the evidence points in the opposite direction: in the canonical texts the emphasis is on the physical and material, and it is Umāsvāti who starts to switch the emphasis to 'internal' action. The change in order is therefore probably not significant. (Devanandin, commenting on Tattvārtha Sūtra 2:25, uses the order vāc, manas, kāya.)
More importantly, Ohira points out that the definition of āsrava as threefold yoga is given for the first time in the Tattvārtha Sūtra (at 6:2).13 (Tattvārtha Sūtra 6:2 can mean that yoga is either the cause of āsrava or that it is itself āsrava; but there is no real ambiguity here, since it is clear that yoga and äsrava are pragmatically synonymous, in the sense that any activity automatically causes the influx of karmic matter.) A further innovation of Umāsvāti's is to classify yoga as śubha ('virtuous' or 'good'), giving rise to punya (merit), or aśubha ('wicked' or 'bad'), giving rise to pāpa (demerit) (Tattvārtha Sūtra. 6:3).14
11 See Ohira p. 61. The identification of Umāsvāti's sources in what follows relies mostly on this work.
12 Ibid. p. 62.
13 Ibid.
14 See ibid. Ohira claims that yoga belongs theoretically to a 'neutral category', but that Umāsvāti reads it in terms of subha-aśubha on the basis of the absence or presence of kasāya (ibid.). However, this analysis seems to be wrong on both counts, since, first, it is Umāsvāti himself who makes yoga into a 'neutral category' precisely through the introduction of the kaşāya doctrine. The underlying feeling of early canonical Jainism, as has been made clear, is that virtually all yoga leads to āsrava and bondage, and is thus ipso facto to be avoided. Only when the kaşāya doctrine (that the binding power of action depends upon the internal state or attitude of the individual) is introduced, is it possible to conceive of yoga as being either binding or not (i.e.

Page #64

50 Harmless Souls
The Sarvārthasiddhi, commenting on Tattvärtha Sutra 6:4, states that karma is of two kinds:
1) samparāyika, leading to samsāra, and 2) iryāpatha, caused by vibrations. Samparāyika is caused by yoga and kaṣāya combined, and iryapatha by yoga alone. However, it is evident that what are being referred to here are not two different types of karma as such, but two different conditions of the jiva whether it has, or is associated with, kaṣāyas or not. Particular karma or karmic matter is not inherently more or
'neutral'). This statement of Ohira's is also wrong in that it claims that Umāsvāti reads yoga in terms of subha and asubha on the basis of the absence or presence of kaṣāya, i.e. that subha yoga leads to iryāpatha karma and aśubha yoga to samparāyika karma. But it is clear that the two categorizations of karma are not synonymous. At TS 8:25, punya karmas, the result of subha yoga, are listed as:
1) sadvedya - pleasure bearing karma 2) subhāyu - good age-karma
3) subha-nāma - good body-making karma
4) subha-gotra - high family-determining karma. (These are divided into forty-two sub-classes of aghatiya or 'nondestructive' karmas.) [On aghātiyā-karmas, see JPP p.132.] All other karmas are papa (TS 8:26).
This list shows that punya karmas are those which determine a good rebirth: So they cannot be synonymous with iryāpatha karma, since that has no karmic effect and is certainly not a cause of rebirth. Rather, both punya and papa karma are forms of sāmparāyika karma. That is to say, both subha and aśubha yoga lead to rebirth of some kind; they are actions motivated by some kind of kaşaya. What is new about Umāsväti's thought here is the idea that there can be such events and categories as subha yoga and punya karma; such a technical division of yoga and karma was inconceivable in the ascetic milieu which is reflected in the doctrines propounded in the earliest parts of the canon. (As we have seen, in the earliest passages virtually all yoga is aśubha and causes papa karman; i.e. all activity binds and so threatens a lower rebirth.) So, for Umasvati, the category of sāmparāyika karma contains punya as well as pāpa karma, a concept which brings the Jaina view of karma into line with the view held by other Indian religions. This, as we shall see, is essentially a lay rather than a monastic or ascetic doctrine. The ascetic is not so much concerned with the possible effects of various types of karma, but with stopping the influx of any kind of karma whatsoever.

Page #65

Umāsvāti's Jainism 51 less binding than any other karma; rather, it is passion which is instrumental in bondage. (A possible explanation for this puzzling terminology will be discussed below.) Thus the emphasis is not on activity as such, but on the accompanying mental or emotional state - on the internal rather than the external.
The reason why 'short-term' (īryāpatha) karma is posited at all, since it has no effect, will emerge later. Here, I suggest that while to begin with virtually all activity caused bondage, for reasons connected with the growing importance of the laity it eventually became necessary to differentiate the relative amount of bondage caused by different actions. 15
How does the Tattvārtha Sūtra's contention that the decisive instrumental factor in bondage is kasāya fit with the main teaching of the earliest canonical texts, namely, that to cause harm (himsā), by any means whatsoever, to any of the innumerable jīva which populate the physical world, is the binding sin par excellence?
At Tattvārtha Sūtra 7:13 (8), himsā is defined as pramattayogāt prāņavyaparopaņam, 'the destruction of life due to an act involving negligence'. 16 The Sarvārthasiddhi comments:
Pramāda connotes passion. The person actuated by passion is pramatta. The activity of such a person is pramatta-yoga.17
Thus the Sarvārthasiddhi differentiates between activity engendered by passion, which results in himsā, and
15 S.A. Jain echoes this historical and institutional divide when he remarks, 'From the real point of view, it is no doubt true that all activities are undesirable, as every kind of activity is the cause of influx and bondage. But from the empirical point of view there is a difference.' - p. 168, fn. 2.
16 Sukhlalji's trans., p. 267.
17 S.A. Jain's trans., pp. 196-197, of: pramādaḥ sakaṣāyatvam tadvānātmapariņāmaḥ pramattaḥ pramattasya yogaḥ pramattayogah.

Page #66

52 Harmless Souls passion-free activity, which does not (i.e. himsā is only himsā because it has been engendered by passion).
This interpretation is clearly tendentious; the technical meaning of pramāda is 'heedlessness', 'carelessness', or 'negligence', derived from vmad + pra, 'to be intoxicated'. 18 Here, therefore, the use of pramāda is comparable to its use in, for instance, the Yoga Sūtras (for example, at 1:30, where pramāda is included in a list of citta-vikṣepas, 'distractions of the mind-field'). And as we have seen, the stress in the earliest canonical sources is on physical 'carefulness', in contrast to the Buddhist concept of appamāda, which has the connotation of 'mindfulness'.19 In the Sarvārthasiddhi , however, the whole emphasis of himsā is shifted on to the internal state of the agent, i.e. on to passion and its effect on the agent: himsā is always himsā to oneself. This goes much further than a literal reading of the sūtra allows; however, as will be seen later, the space for such an internalised doctrine of bondage is cleared by Umāsvāti in the Tattvārtha Sūtra, if not completely developed there. The Sarvārthasiddhi simply takes Umāsvāti's thought a step further. Mere injury, according to the Sarvārthasiddhi, even killing, does not stain one with the sin of himsā; i.e. it is not himsā as such, and does not bind.
Apparently quoting from Kundakunda's Pravacanasāra, Pūjyapāda (Devanandin), in his commentary on Tattvārtha Sūtra 7:13, affirms that:
When a monk goes on foot with carefulness, sometimes small insects get crushed under his feet and die. Still there is not the
18 Cf. TS 7:4, where the five observances of the vow of ahimsă are all couched in terms of 'carefulness'; and the SS itself (on TS 8:1) defines pramāda as 'misinterpreting' injunctions, and indifference in relation to kriyā (action).
19 See, for instance, the Buddha's last exhortation (Dīgha Nikāya 16): ... appamadena sampadetha - 'strive diligently'; also see Buddhist Dictionary p. 22: 'In the commentaries (appamāda) is often explained as the presence (lit. "non-absence") of mindfulness (satiyā avippavāsa)'.

Page #67

Umāsvāti's Jainism 53 slightest bondage of sin in this case. ['Carefulness' being the opposite of pramāda or 'negligence'.]20
The emphasis here is clearly on the state - what amounts to the mental condition - of 'carefulness', not on the physical injury as such. This is confirmed by Pūjyapāda when he goes on to express the converse argument that a 'mere passionate attitude even without the severance of vitalities constitutes violence'.21
Again the Sarvārthasiddhi quotes a verse that also occurs in the Pravacanasāra which summarises this line of argument:
He who acts with negligence commits injury whether death is caused to organisms or not. And he who proceeds with proper care
20 S.A. Jain's trans., p. 197, of: uccālidamhi pāde iriyāsamidassa niggamatthāne | āvādejja kulimgo marejja tajjogamāsejja ||
nahi tassa tannimitto bamdho suhumovi desido samaye || This is to be found in Jayasena's recension of the Pravacanasāra at 3:17 (1-2), Upadhye's edition (see Appendix 3, Pravacanasāra 3.17b). Given the compilatory nature of the Pravacanasāra (see below), it cannot be certain that such quotations originate there. The attribution is made by P.S. Shastry (ed. SS, Banaras 1955), but it is possible, even likely, that Pūjyapāda and Kundakunda are quoting from a common source. Indeed, Upadhye in his translation of Pravacanasāra 3;17 (1-2) assumes that this is a quotation without identifying the source. In fact, Upadhye speculates that these are very old traditional gāthās, belonging to both Digambaras and Svetāmbaras (1935, p. liiiff.). Moreover, the fact that Pūjyapāda also quotes here a verse which appears as Pravacanasāra 3:16 in Amrtacandra's recension (see below), indicates that he had access either to an earlier version of the Pravacanasāra than Amstacandra's (the earliest we possess), i.e. the one from which Jayasena drew as well, or that he is quoting directly from Kundakunda's source. (See also Viy. XVIII.7.1, where the same idea is presented in a more technical way.) It is therefore not desirable to draw conclusions about the dating of the Pravacanasāra from this coincidence.
21 S.A. Jain's trans., p. 197, of: prāņavyaparopaņābhāve 'pi pramattayogamātrād eva himsesyate - SS on TS 7:13.

Page #68

54 Harmless Souls
does not contract bondage of karma by mere injury.22 Pūjyapāda concludes:
He who has passions causes injury to himself by himself. Whether injury is then caused to other living beings or not, is immaterial.23
This argument is derived from the assertion that himsā, the activity which binds, is actually produced by passions. Similarly, P.S. Jaini, using the Sarvārthasiddhi as his. source, notes that the subtlest forms of the passions are called samjvalana ('the smouldering'). These are not sufficiently strong to prevent one entering the mendicant's path, but they induce 'an insidious state of apathy or inertia (pramāda), a lack of drive with regard to the actual purificatory practices enatailed by that path'.24
To summarise, Pūjyapāda is claiming that 'negligence' (pramāda), and thus himsā, is only produced when passions are involved. Consequently, bondage can only occur when there is some kind of volitional activity motivated by passion.25 This is essentially in agreement with the definitions of bondage given by Umāsvāti at Tattvārtha Sutra 8:
Wrong belief, non-restraint, carelessness, passions and
22 S.A. Jain's trans., p. 197, of: maradu va jiyadu va jivo ayadācārassa nicchidā himsă | payadassa natthi bamdho himsāmitteņa samidassa 11
Pravacanasāra 3:17 || See Pravacanasāra 3:16; and see p.156ff., below, for a further discussion of this gāthā.
23 S.A. Jain's trans., p. 197 of: svayam evātmanā "tmānam hinasty ātmā pramādavān
pūrvam prānyantarānām tu paścāt syād vā na vā vadhaḥ || This is in fact a quotation from a Sanskrit source which I have not been able to identify
24 JPP pp. 120. See SS on TS 8:9; cf. TS 10:1. 25 See JPP pp. 112-113.

Page #69

Umāsvāti's Jainism 55 activities are the causes of bondage. [8:1]
Because of its connection with passion, a soul takes on particles of matter liable to become karma. This is bondage. [8:2 (2/3)]
Bondage is of four kinds, according to the nature of karmic matter (prakrti), the duration of karma (sthiti), the intensity of the fruition of karma (anubhāva), and the quantity of space-points of karma (pradeśa). [8:3 (4)]26
According to the commentaries, prakrti and pradeśa are caused by yoga, sthiti and anubhāva by kaşāya. It is yoga which attracts karma to the soul in the first place, deciding its type and quantity, and kasāya which causes it to adhere, deciding its duration and intensity. Thus, according to Tattvārtha Sūtra 8:2, it is kasāya which underlies all bondage (i.e. it is the cause of all himsā and thus the cause of all pramāda as well).
However, in the list of the five causes of bondage given at Tattvārtha Sūtra 8:1, kaşāya and pramāda are listed as independent causes of bondage.
According to a modern commentator, there are three traditions regarding the number of the causes of bondage: viz. there are either five causes, as given above, or four (pramāda is excluded), or there are just two, kaşāya and yoga.27 But these traditions can be harmonised by taking pramāda as a type of either avirati or kaşāya, and then by viewing mithyadarśana and avirati as not essentially different from kaşāya. This leaves kasāya and yoga as the only distinct causes of bondage, which is Umāsvāti's conclusion.
In some canonical texts the five causes of bondage
26 mithyādarśanāviratipramādakasāyayogā bandhahetavaḥ
[TS 8:1] sakaṣāyatvāj jīvaḥ karmaṇo yogyān pudgalān adatte sa bandhah [TS 8:2 (2/3)]
prakstisthityanubhāvapradeśās tadvidhayaḥ [TS 8:3 (4)] 27 Sukhlalji's trans. and commentary on TS, p. 298f.

Page #70

56 Harmless Souls
(bandhahetavaḥ), enumerated at Tattvärtha Sutra 8:1, occur as asravadvāras - causes of (literally, 'entrances for') asrava.28 As Ohira points out in connection with this, theoretically there is no difference between the root causes of asrava and of bandha,29 but Umāsvāti took yoga to be the root cause of asrava (on the basis that the threefold yogas are present in all the other causes), and then classified yoga as causing both īryāpatha āsrava and sāmparāyika āsrava while placing the rest of the canonical āsravadvāras (viz. mithyādarśana, avirati, pramāda, and kaṣāya) in the category of samparāyika āsrava only. The crucial alteration here, however, is Umāsvāti's division of yoga into that which stems from kaṣāya and involves mithyādarśana, avirati, and pramāda, and that which is free from kaṣāya, i.e. it is only one kind of yoga which causes bondage - the other (passionless) kind does not bind. Thus Umāsvati has in effect made a distinction between the causes of asrava and the causes of bondage: whereas both kinds of yoga (passionless and passionate) cause asrava, only one kind causes bondage.
The canonical idea that mithyādarśana, avirati, pramāda, kaṣāya, and yoga, as the causes of asrava, are indistinguishable from the causes of bondage, makes it clear that before Umāsvāti there is no technically formulated conception of any kind of asrava which does not bind.30
At Tattvärtha Sutra 8:9 (10) the four kaṣāya are included in a list of twenty-eight deluding karmas (as caritra-mohaniya, 'right-conduct deluding karmas'), but they are not mentioned there as the specific cause of karmic matter adhering to the jiva. It is clear that Umāsvāti is here
28 See, for example, Sthāna 5.2.517, and Samavāya 16, cited by Ohira, p. 62.
29 Ibid.
30 See the discussion, p. 14, above, of the ancient meaning of asrava according to Alsdorf (1965), indicating that asrava was probably originally synonymous with bondage.

Page #71

Umāsvāti's Jainism 57 taking over an earlier list without any attempt to integrate it with later doctrines.31 That is to say, the four kaşāya are functioning here as they did before Umāsvāti collated them with the principle of instrumentality. The relevant sourcepassage from the Uttarajjhayana makes this clear:
The two kinds of mohaniya referring to conduct are: 1) what is experienced in the form of the four cardinal passions (kaşāya); 2) what is experienced in the form of feelings different from thein (no-kaṣāya or 'subsidiary passions'). 32
According to the Sarvārthasiddhi, the cāritra-mohanīya karmas both cause the passions and are caused by them (i.e. the passions are at the same time both the result of delusive conduct and what lead to further delusive conduct).33 But even within the circularity of this later commentarial gloss, it is clear that the passions are understood as leading to bondage only through the indirect route of conduct which is deluded - i.e. activity which is harmful in some way.
2.3 Sāmparāyika and iryāpatha karma in the Tattvārtha
Sutra
As has been seen above, 34 yoga accompanied by kaşāya gives rise to the influx of sāmparāyika karma, i.e. karma which binds and thus leads to rebirth - literally, it is the 'passage to the other world' (samparāya). At Tattvārtha Sūtra 6:5 (6), Umāsvāti subdivides sämparāyika karma into four types, corresponding to the five causes of bondage at Tattvārtha Sūtra 8:1. Thus sāmparāyika karma is caused
31 In fact this passage is probably directly derived from Utt. 33 - see Ohira p. 64, and below.
32 Based on Jacobi's trans. (1895, p.194) of Utt. 33.10: cărittamohanam kammam, duviham tu viyāhiyam | kasāyamohanijjam tu nokasāyam taheva ya || 10 || 33 See SS on TS 8:9 (10) and S.A. Jain's trans., p. 223-224. 34 See above, p. 47ff.

,

Page #72

58 Harmless Souls by the activity of:
1) the 5 senses (indriya) - touch, taste, smell, sight, hearing, 2) the 4 passions (kaşāya) - anger, pride, deceitfulness, greed, 3) the non-observance of the 5 vows (avrata), 4) the 25 activities (kriyā).
These correspond to the 5 causes of bondage (Tattvārtha Sūtra 8:1) in the following way:
1) the 5 indriyas correspond to pramāda (indriya, as Ohira points
out, is explained in the bhāsya on 6:5 (6) as pañca
pramattasyendriyāņi),35 2) the 4 kaṣāyas correspond to kaṣāya, 3) avrata corresponds to avirati, 4) the 25 kriyas correspond to mithyādarśana (mithyādarśana is
included as the twenty-fourth of the twenty-five kriyās, and mithyātva occurs as the second of the twenty-five).36
The fifth cause of bondage, yoga, is, as has been shown, defined by the other four (i.e. it is yoga with 'passion'), and it is activity (yoga) in the four categories enumerated which gives rise to sāmparāyika karma.
Comparing these two classifications, the question arises why it is that, when he comes to list the causes of sāmparāyika karma (Tattvārtha Sūtra 6:5 (6) = the 5 causes of bondage (Tattvārtha Sūtra 8:1]), Umāsvāti substitutes a list of 25 kriyā for mithyādarśana, which is only one item on that list.
According to Ohira, kriyā is repeatedly propounded in the early canonical works as the cause directly inviting āsrava, 'so Umāsvāti must have wanted to lay emphasis on it by counting twenty-five in all in place of mithyātva which is just part of them.37 But as Ohira herself has
35 Ohira p. 62. 36 See J.L. Jaini's list in his trans. of TS (1920) p. 126. 37 Ohira pp. 62-63.

Page #73

Umāsvāti's Jainism 59 already pointed out,38 the inventory of kriyā 'has been worked out independently in the long Agamic period', so this list is not Umāsväti's invention. Rather, he is trying to integrate canonical material into his own division between binding and non-binding yoga - yoga with and without kaṣāya.39 This process is revealed when the list of 25 kriyā is examined in detail, for not all the activities listed are compatible with their classification as giving rise to sāmparāyika āsrava and further bondage in samsāra.
The fourth kriya in the list is given as 'īryāpatha'.40 The term 'īryāpatha' has first occurred in this context in the previous aphorism (6:4 [5]) as the kind of asrava which does not lead to transmigration, viz. that kind experienced by persons free from passions (in opposition to sāmparāyika āsrava, it gives rise to short-term karma). What then is its meaning at Tattvārtha Sūtra 6:5 (6), where it appears among the list of kriya which give rise to sāmparāyika karma? If the meaning is the same, the two sūtras contradict each other.
İryapatha is derived from the root ir (II), meaning 'to go', 'to move'. Thus the Sarvārthasiddhi on Tattvärtha Sūtra 6:4 (5) comments, 'Iraṇam means iryā, yogo or movement (vibrations). That karma which is caused by vibrations is called iryapatha',41 i.e. it is that (non-binding) karma which is caused by yoga alone, as opposed to that (binding) karma which is combined with passions (kaṣāya). Similarly, Sukhlalji comments:
The reason why this karma of a duration of one samaya is called iryāpathika is that in the absence of all passion it is bound down merely through the patha or instrumentality of īryā or acts like
38 Ibid. p. 62.
39 On kiriyā / kriya, cf. p. 40ff., above.
40 See J.L. Jaini, (TS 1920) p. 125, on TS 6:5. S.A. Jain's trans., p. 169, of:
41
iraṇam īrya yogo gatir ity arthaḥ | taddvārakam karma iryāpatham.

Page #74

60 Harmless Souls
coming and going. 42
However, when he comes to deal with iryāpatha-kriyā at Tattvārtha Sūtra 6:5 (6), Sukhlalji glosses it in exactly the same way, as 'kriyā which causes either the bondage or the experiencing of iryāpatha karma - that is, karma of a duration of one samaya'. Such a reading leads to the incompatibility between the two sūtras outlined above (i.e. how can this be included in a list of sāmparāyika karma / āsrava, when it is apparently of the other type of āsrava / karma - īryāpatha?). Sukhlalji is aware of the problem, but is unable to resolve it. He comments on the exceptional position of īryāpatha kriyā on this list:
Of the kriyās ...(mentioned above)... there is only one - viz. īryāpatha kriyā - that is not āsrava for a sāmparāyika karma: as for the remaining ones since they are all impelled by kaşāya they are all cause-of-bondage for sāmparāyika karma. And when all these kriyās are here called āsrava for a sāmparāyika karma that is done simply because most of them (really all of them except īryapathiki) are in fact so.43
In other words, he can offer no explanation at all. But this does highlight the problem, and indicates that the list of kriyā was probably taken over in its entirety from an earlier source which does not fit the later division into sāmparāyika and iryāpatha āsrava.
In the context of a list of kriyā , the correct interpretation of Tattvārtha Sūtra 6:5 (6) is probably that of the Sarvārthasiddhi , viz. 'īryāpathakriyā is walking carefully, by looking on the ground (for living beings which may be trodden and injured)'. 44 But this in turn raises two problems: 1) why does the same term apparently
42 Sukhlalji (TS 1974), p. 233. 43Ibid. p. 236. 44 S.A. Jain's trans., p. 170 (cf. J.L. Jain (1920) p. 125) of: īryāpathanimitteryāpathakriyā - SS on TS 6:5.

Page #75

Umāsvāti's Jainism 61 have two different technical meanings? 2) what does Tryāpatha kriyā ('walking carefully') have to do with kaşāya and bondage? (How can 'care in walking' be said to be characteristic of a jīva which has passions?)45
Suhklalji attempts to solve the first problem by running the two meanings together (as discussed above), but at the cost of making the two sūtras incompatible. The Sarvārthasiddhi's commentary on Tattvārtha Sūtra 6:5 (6), that the 5 senses, the 4 passions, the 5 kinds of vowlessness, are the causes of influx, and that the 25 kinds of activity (kriyā ) are the effects, similarly fails to remove the incompatibility between the two sūtras.46 Again iryāpathakriyā as defined (and samyaktva kriyā ) is out of place in such a list. One incoherence is replaced by another.
This incompatibility of Tattvārtha Sūtra 6:4 (5) with 6:5 (6) indicates that Umāsvāti is attempting to run together two different categories or lists, one developed later than the other (i.e. one containing the term 'īryāpatha' as used at an earlier date and with a different meaning). In other words, Umāsvāti is taking over lists of kriyā from earlier sources - as Dixit remarks, all four categories of activities listed at Tattvārtha Sūtra 6:5 (6) were in 'more or less extensive use' independently of each other in the texts available to Umāsvāti47 - without attempting to make them fit his definition of sāmparāyika ásrava (that pertaining to persons with passions), which he merely superimposes.
It is interesting to note that at Tattvārtha Sūtra 9:5 iryāsamiti, 'proper care in walking', is named as one of the
45 There is a similar problem in explaining the presence of samyaktva kriyā - 'that which strengthens right belief - in the list of kriyā (no.1). The other 23 kriyā are at least negative, although they are by no means obvious characteristics of the 'passionate'.
46 See S.A. Jain p. 171: etānīndriyādīni kāryakāraṇabhedādbhedamāpadyamānāni - SS on TS 6:5.
47 Dixit in the preface to Sukhlalji (TS 1974) pp. 7-8.

Page #76

62 Harmless Souls five types of samiti ('carefulness' or 'awareness'),48 and that samiti is listed at Tattvārtha Sūtra 9:2 as one of the means to samvara. The latter is defined at Tattvārtha Sūtra 9:1 as āsravanirodhaḥ 'the stoppage of inflow' - presumably of that karmic matter which binds, since all activity causes āsrava as such. Thus the types of samiti clearly cannot belong to the category of sāmparāyika karma / āsrava. This provides further evidence that the linkage of Tattvārtha Sūtra 6:4 and 6:5 by Umāsvāti is incoherent, and that the original meaning of iryāpatha was 'care in walking', and only at a later date acquired the general meaning of 'short-term karma'.
It is perhaps significant that in Buddhism iryāpatha (Pali: iryāpatha) is more or less a technical term for 'the four postures' - walking, standing, sitting and lying.49 If we apply this to the Jain case, it is easy to see that iryāpathakriyā would be care in all bodily movements, since the four postures cover all eventualities. Of the four, walking would of course be the most dangerous, but the others are presumably included as moments when a monk might easily do harm. Consequently, as īryākriyā refers to all an ascetic's movements, then it covers all the physical harm he might do, and thus it is responsible for the totality of his karma.
From this it is possible to see how iryāpatha's change in meaning may have taken place. 'Care in walking', a term which includes all controlled ascetic activities, causes so little himsā (although some is inevitable given the distribution of souls), and gives rise to such slight āsrava, that the quantity of karma accrued in this way is small enough to be shed in a single lifetime, and thus it does not lead to another birth (sāmparāyika). A distinction is therefore made between activities which lead to the influx of long-term karma, which is binding, and
48 See p. 69, above, for the other kinds of samiti.
49 See Buddhist Dictionary p. 81. They form a subject of contemplation and an exercise in mindfulness in the Satipatthāna Sutta.

Page #77

63
Umāsvāti's Jainism those which lead to an influx of short-term karma, which is not. Such activities are all related directly to himsā and non-himsā, and especially to physical activity. That is to say, the behaviour of the ascetic is being distinguished absolutely from that of the non-ascetic or lay-person in terms of soteriological consequences.50
There are, therefore, two types of asrava: that which occurs to jivas in lay bodies, and that which occurs to those in the bodies of monks. The difference arises from the difference in behaviour between ascetics and others. Similarly, iryāpatha and sāmparāyika are, strictly speaking, two different quantities of karma, one accrued by mendicants and the other by householders. But the distinction is considered to be so fundamental that they become in effect two different types: short-term, and that which leads to a further rebirth. And in these soteriological terms, the division is absolute.
In this way iryāpatha acquires the general meaning of 'short-term karma'. At a later stage, it is given a specific technical definition, by Umāsvāti, as that influx of karma which is short-term because it is free from passion; it is therefore no longer directly linked to the amount of (physical) himsă caused, but to internal states, motivation, etc. In other words, Umāsväti's definition of himsā - 'Injury is the destruction of life out of passion'51 - bears witness to the development of a less exclusive, i.e. more laycompatible ethic in the intervening period.52 If this modei is correct, then the concept of iryapatha karma developed in the period when asrava and the resulting bondage of the jiva by karmic matter were still seen as the results of physical activity alone. In that case the instrumentality of kaṣāya would be a later accretion, based upon the
50
51
See the passages quoted from the Viy., etc., p. 40ff., above. pramattayogāt prāṇavyaparopaṇam himsā - TS 7:13 (8) and
SS.
52 Cf. SS on TS 7:22, where TS 7:13 is quoted as a justification of lay sallekhana.

Page #78

64
Harmless Souls
supposition that 'passion' or 'intention' can to some extent be controlled by someone who goes about his worldly business (i.e. it is a matter of attitude), whereas restraint of physical activity demands a particular kind of 'extraworldly' ascetic discipline. And as will now be made clear, the practice of monks and nuns remains unaffected by this internalisation, which effectively comes to function as a means of giving the laity a theoretical foothold on the path to salvation.
2.4 'Activity' in the Tattvärtha Sutra
According to Tattvārtha Sūtra 6:7, jīva (what is sentient, i.e. the soul) and ajīva (what is insentient) constitute the adhikaraṇāḥ - the 'substrata' - of influx; in other words, the causes of influx. Or as the Sarvārthasiddhi puts it, being adhikarana is 'the condition of being the instruments of injury and so on', and thus the condition of being instrumental in the influx of binding karmic matter.53 Commenting on this, Sukhlalji says that 'both jiva and ajīva are called adhikarana - that is to say, a means, implement or weapon of karmic bondage'.54 This seems to make 'karmic bondage' the active principle which needs the adhikarana in order to express itself, rather than the jiva being the active principle. But by the very fact of being available for āsrava, the jīva is in a sense instrumental in its own bondage. Karmic bondage can only become 'active' if the jiva behaves in certain ways (enumerated at Tattvärtha Sūtra 6:8), and it is in this sense that the jiva is the 'substratum' - a reading which is compatible with the usual meaning of adhikaraṇa as 'that in which anything happens'.
The ways in which a jīva can cause injury are numbered as 108, as follows:
The substratum of the living is planning to commit violence,
53 S.A Jain's trans., p. 172, of: himsādyupakaraṇabhāva. 54 Sukhlalji's commentary on TS, p. 239.

Page #79

Umāsvāti's Jainism 65 preparation for it and commmencement of it, by activity, doing, causing it to be done, and approval of it, and issuing from the passions, which are three, three, three and four respectively. [Tattvārtha Sūtra 6:8 (9)]55 (These are the causes of
the influx of karma in general; specific types of karma · have, of course, the same general causes but also have specific causes which are subtypes of the general causes.)56
With the help of the commentaries,57 Tattvārtha Sūtra 6:8 (9) can be presented in a schematic form (see Table). The notably new thing about this schema or formulation, compared with those associated with its component parts (which, as we have seen, are scattered throughout the early Svetāmbara canon), is that whichever way one reads it, left to right or right to left, passion (kasāya) is instrumental in causing violence, and thus äsrava and bondage. Note that nothing is said here of action without passion; it is clear that for Umāsvāti ārambha means premeditated, violent action (himsā). In other words, the action which binds is violent action engendered by passions; from which it may be inferred that passionless, non-violent action does not bind (although it does cause an influx of matter into the soul in the form of iryāpatha karma). And it is clear from Tattvārtha Sūtra 6:4-6 that this enumeration of the 108 ways in which influx can be caused to a jīva refers specifically to the person who is actuated by passions (sakaṣāya) (Tattvārtha Sūtra 6:4).
This is further clarified by the Sarvārthasiddhi on Tattvārtha Sūtra 6:8 which, taking for its example 'bodily impulsion' (kāya-samrambha), i.e. the determination to do violence through bodily activity (one of the three yogas),
555. A. Jain's trans., p. 172, fn. 1, with minor alterations of: adyam samrambhasamārambhārambhayogakrtakāritānumatakasāyavisesais tristristriścatuścaikaśaḥ [TS 6:8 (9)]. 56 See TS 6:10f. 57 See SS, J.L.Jaini (1920), Sukhlalji (TS 1974).

Page #80

1
66 Harmless Souls subdivides it as follows:
Bodily impulsion performed (kṛta) by anger (krodha) Bodily impulsion performed by pride (māna) Bodily impulsion made by deceitfulness (māyā) Bodily impulsion made by greed (lobha) (i.e. all done by oneself)
Bodily impulsion instigated (kārita) by anger by pride
by deceitfulness
by greed
(i.e. done by one's agent)
Bodily impulsion approved (anumata) by anger
by pride by deceitfulness by greed
(i.e. done by others with one's approval).
It is also interesting to note that it is no longer simply the act of doing violence which is binding, but also the impulsion or intention to do it, and the preparation for it (samrambha and samarambha) (Tattvārtha Sutra 6:8). This threefold division of himsā may be seen as the logical corollary of, or complement to, the development of the idea of the instrumentality of kaṣāya as the binding agent, for passion may be as strong or stronger in intention and preparation as in outcome (actual violence).58
Tattvārtha Sūtra 6:9 (10) deals with the adhikaraṇa of

58 These three- samrambha, samārambha, and ārambha - are also to be found at, for instance, Utt. 24.19ff., although their meaning there may be somewhat different if Jacobi's trans. (1895, p. 135) is followed. Schubring quotes the commentary which understands the terms to mean - as in the TS - 'two stages of preparation and the performance of forbidden thinking, speaking and acting' (1962, para. 173 / p. 304). But again there is no connection with a technical mechanism of bondage such as Umāsvāti's kaṣāya doctrine.

Page #81

Umāsvāti's Jainism 67 the ajīva type. It would seem to refer to pudgala (matter) which can in some way aid, or be manipulated by the individual to bring about karmic influx. This substratum is divided into:
2 (kinds of) nirvartanā ('production' / 'performance'):
i) mūlaguna - of body, speech, mind, inhalation, exhalation, ii) uttaraguņa - making objects of wood, clay, etc., pictures
and statues. (That is to say, the mūlagunas are what the body performs naturally, and the uttaraguņas are what the body does by extension.)
4 (kinds of) niksepa ('placing' / 'putting down');
i) apratyavekșita - 'without seeing' ii) duspramārjita - 'without cleansing of dust, etc.' iii) sahasa - "hurriedly' iv) anābhoga - 'inattentively/putting something where it ought
not to be put'.
2 (kinds of) samyoga ('combining' / 'mixing up'):
i) bhaktapāna - 'food and drink'/ different foodstuffs' ii) upakarana - 'mixing up things - implements, clothes, etc.'
3 (kinds of) nisarga ('movement'/ 'urging'/ 'operating'):
i) kāya - 'by body ii) vacana - 'by speech' iii) manas - 'by mind'
[Definitions taken from Sarvārthasiddhi.]
As Ohira has pointed out, the materials used in Ch. 6 of the Tattvārtha Sūtra are widely dispersed throughout the canonical texts.59 And it would seem that in dividing the substratum (adhikaraņa) of influx into two, jīva and ajīva, Umāsvāti is either drawing on two different canonical lists or, more likely, in making this division he is expanding his
59 Ohira p. 61.

Page #82

68 Harmless Souls own kaşāya doctrine in the first adhikaraņa - that of jīva - and merely repeating the received (ascetic) adhikarana of influx in the ajīva list. In other words, the division of the mechanism of bondage into two types of karma, that which causes bondage - which 'sticks' or adheres - and that which does not, is prepared for or justified by this division into two types of adhikarana ; mere physical action of a mechanical kind, and motivated, impassioned action, are separated. Such a separation would not be necessary unless the soteriological consequences of there being two substrata of influx (i.e. two types of karma producing action) were perceived as different. Both can bind, but Umāsvāti, in attempting to combine the two, puts emphasis on the first: it becomes a question of behaviour and attitude, not just of behaviour.
This opens the way for the later commentators and their preoccupation with attitude. Furthermore, there is no logical reason why there should be influx of two types of karma - binding and non-binding - unless the kasāya doctrine is a later addition. For if the idea that it is only passionate action which binds had been there in the earliest form of the doctrine, there would have been no reason to posit a complex doctrine of 'stickiness', etc.: Jains would simply have been able to say that activity accompanied by passion causes influx of karma and other activity does not. (The original doctrine must have been that physical activity alone was the source of karma and thus binding.) It is clear, therefore, that there are two layers of doctrine here, and that the superimposition of one upon the other marks a change for which there are historical rather than logical (i.e. strictly doctrinal) reasons.
Returning to the adhikaraņa of the ajīva type enumerated at Tattvärtha Sūtra 6:9 (10), it is probable that this list is based upon the original rules for ascetics in their wandering life. In fact all the categories are to do with himsā caused by physical activity, and relate very closely to the 5 samiti. These are enumerated at Tattvārtha Sūtra 9:5, where they are listed as one category of the modes of

Page #83

Umāsvāti's Jainism 69 behaviour which lead to samvara, the cessation of influx (Tattvārtha Sūtra 9:2).60
The 5 samiti are:
i) īryā-samiti - care (awareness) in walking, ii) bhāṣā-samiti - care in speaking, iii) esanā-samiti - care in accepting alms, iv) ādāna-nikṣepaņa-samiti - care in picking things up and in
putting them down, v) utsarga-samiti - care in performing excretory functions. 61
Comparing this list with that of the four divisions of the adhikaraņa of the ajiva type, given above, the following correspondences can be discerned:
1) īryā- and bhāsā-samiti correspond to the 2 nirvartanā, 2) eșaņā-samiti corresponds to samyoga, 3) ādāna-nikṣepana-samiti corresponds to niksepa, 4) utsarga-samiti corresponds to nisarga.
That is to say, the 5 samiti are the antidotes to the 4 adhikaraņa of the ajīva type: they advocate care in relation
60 It is worth noting that, according to Ohira p. 65, samvara is not defined in the canonical texts in the fashion expressed by Umāsvāti at TS 9:1 (as the stoppage of the inflow of karmic matter into the soul - āsravanirodhaḥ samvarah). Nor do the sixfold samvaradvāras (TS 9:2) occur as a set category in the canon; rather, they were formulated by Umāsvāti. They are, nevertheless, predominantly physical in character. The first among them is gupti, defined at TS 9:4 as samyagyoganigraho gupti - 'restraint / prevention is proper control (nigraha) over yoga' (i.e. over the activities of body, speech and mind). Thus it is still restraint of activity which is seen as the pre-eminent means to release (as opposed to attaining a better re-birth), rather than restraint of 'passions' as such. Also note that gupti here, as restraint / control of the three yogas, would appear to have the meaning normally assigned to 'yoga' in other schools.
61 Cf. Pāli sati, the 'mindfulness' or 'awareness' of early Buddhism. Samiti may be a backformation from this, therefore having the sense of 'awareness'.

Page #84

70 Harmless Souls to matter. Taking each of these in turn:
1) iryā-samiti and bhāṣā-samiti refer respectively to care in walking and care in speaking. The connection of these with the ajīva-adhikaraṇa nirvartanā ('performance'), which is divided into mūlaguna and uttaraguna, seems at first to be only partial. That is to say, the connection holds in so far as, if mūlaguna refers to what the body performs or does naturally, simply by being the body, and uttaraguna to what it does by extension, then iryā and bhāṣā-samiti correspond to the first two mūlaguņa, the 'performance of body and speech i.e. one should take care in walking and speaking). This would leave the mūlaguņa of mind, inhalation and exhalation, and the uttaraguņa as later additions to, or elaborations of, the original doctrine, probably to meet with circumstances previously unforeseen (e.g. the manufacture of images, etc.). ('Mind' may well have been added simply because the formula 'body, speech, and mind' had become a cliché.)62
2) eșaņā-samiti refers to care in accepting alms, and clearly corresponds to the ajīva adhikarana samyoga, although the latter has the extended sense of mixing up or (literally) contact with implements, clothes, etc. (upakaraņa), as well as mixing up or contact with food and drink (bhaktapāna).
3) ādāna-niksepana-samiti corresponds to the ajīva adhikaraṇa niksepaņa, again elaborated into four ways of putting things down carelessly.
4) utsarga-samiti is the rule of conduct which
62 Note that mūlaguna in this context should not be confused with the '8 basic restraints' prescribed for the Jaina layperson, also known as mūlaguņa, viz. abstaining from partaking of meat, alcohol, honey, or any of five kinds of figs (see JPP p. 167). This definition of mūlaguna does not occur in the TS.

Page #85

Umāsvāti's Jainism 71 prescribes care in performing excretory functions. The ajīva adhikarana nisarga is defined by the commentators as 'urging', 'movement', and 'operating!.63 Thus, according to the Sarvārthasiddhi, 'Urging (behaviour) is of three kinds, urging the body, speech and mind to act'.64 It should be noted, however, that this threefold division is not made by Umāsvāti himself, and that nisarga as well as utsarga can have the meaning 'evacuation of excrement'. This may indicate that Umāsvāti's commentators (and possibly Umāsvāti himself) had the samvaradvāra gupti in mind when they defined the ajīva adhikaraṇa nisarga in this way. Gupti is defined (at Tattvārtha Sūtra 9:4) as samyagyoganigrahaḥ, proper control (nigraha) over activity (yoga) - i.e. control over the activities of body, speech, and mind, which would be the antidote to urging the body, speech, and mind to act (the three kinds of nisarga).
The likelihood that nisarga originally referred to evacuation of excrement is strengthened by its incompatibility (as it is defined by the commentators) with the other adhikaraņa-ajīva. For the presence of manas, regarded as denoting the operation of the mind of an individual, introduces the idea of intention, of internal 'action' into a list which refers to the physical manipulation of matter (pudgala).65
It is possible to conclude, therefore, that of Umāsvāti's categories ajīva and jīva adhikaraņa, the former was derived from the original monastic rules concerned with preventing the influx of binding karma, since it is occupied
63 S.A. Jain p.174; J.L. Jaini (1920); Sukhlalji (TS 1974).
64 S.A. Jain's trans. p. 174 of SS on TS 6:9: nisargas trividhaḥ - kāyanisargādhikaraṇam vāg... mano..., etc.
65 Such incompatibility is, of course, reduced if yoga is read in the sense suggested by Dixit (see above) - viz, as doing an act oneself, having it done through one's agent, or allowing it to be done by someone else; but that is not the way the commentators take it.

Page #86

72 Harmless Souls entirely with physical activity and the himsā caused by it. In other words, the list of ajīva-adhikaraṇa is based on the canonical rules for those wandering ascetics whose primary concern was not with a particular attitude of mind, or intention, but with the avoidance of any physical action which might cause harm to any of the myriad jīvas by which they were surrounded. The latter (jiva-adhikarana) is added by Umāsvāti to account for the proliferation of karmic positions that less than perfect ascetics, or lay people, can find themselves in.
2.5. Parigraha: the householder and the kaṣāya doctrine
In the developed kaşāya doctrine (the one which is presented by Umāsvāti) what was once a single instrument of bondage (viz. violent or harming activity engendered by lobha and parigraha) has now become two: bondage is caused by passion (kaşāya) and harming activity. It is no longer simply the case that passions, or lobha, etc., lead to violent activity, rather violent activity is now not even considered 'violent' unless it is accompanied by passion; and thus, without passion, it is no longer binding either. The notion, expressed in the earliest textual layers, that even accidental himsā is binding, has been removed. How is this development, which is at the heart of the tension between later doctrine and practice, to be explained?
As has been shown, in the earliest texts parigraha or lobha is considered, along with violence (ārambha), to be the worst kind of behaviour, and such parigraha or 'greed' is inextricable from the life of the householder. Indeed, for the composers of monastic texts such as the Dasaveyāliya Sutta, parigraha is precisely what defines the state of being a householder. Thus, according to these sources, there is very little possibility other than a bad rebirth for a householder. The householder's state is one of parigraha, and thus of ārambha, simply because he is a householder;

Page #87

Umāsvāti's Jainism 73 only the monk or nun has the potential to cease from parigraha and ārambha, and thus attain mokşa. The idea that acts involving the employment of human and animal labour were particularly sinful, and others less so, does not appear in the earliest doctrinal layer. There, all acts which harm jīvas are considered to be effectively binding. Indeed, the householder only puts in an appearance in these texts to act as a kind of lighthouse, warning the ascetic away from, or at least setting severe limits to, social contact. The householder's inevitable rebirth and suffering are pointed out, like wreckages, as a warning to those who come too close or are tempted to return to life in the world.
In later Jaina theory, however, the pious householder is considered to be on the same soteriological continuum as the monk: the former may take partial vows (aņuvrata) which are seen as preparing him or her for the eventual assumption of the mahāvrata of the mendicant.66 Thus while the aņuvrata of ahimsā, which applies only to trasa beings, can never be soteriologically sufficient - it cannot lead in itself to liberation - it is nevertheless a step upwards on the ladder which leads to mokşa; one can expect at least a better rebirth. Such a progression is never contemplated in the very earliest textual layers, where the idea of a 'pious' householder is not even admitted. Indeed, such a concept would be a contradiction in terms.
Moreover, according to later Jaina doctrine, parigraha is defined as mūrcchā,67 'infatuation' or the 'delusion of possession'. 68 In the Sarvārthasiddhi mūrcchā is glossed as 'not turning away from the aims of acquiring and keeping conscious and unconscious externals, such as cows, buffaloes, jewels, pearls, possessions, etc., and
66 See JPP p. 160ff. 67 TS 7:17. 68 JPP p. 177.

Page #88

74 Harmless Souls internal conditions, such as desire, etc.'.69 The Sarvārthasiddhi goes on to say, delusion 'is the root of all imperfections. When someone has the idea "This is mine", the need to take care of it, etc., (also) arises. And from that violence necessarily follows'. 70 And as P.S. Jaini points out, the term parigraha is 'further made synonymous with the four passions (kaşāya) and nine sentiments (no-kaṣāya) ...; these are known as the "internal possessions" and their renunciation (the avoidance of activities which generate them) constitutes the essence of the aparigrahavrata' (the fifth anuvrata of the layman according to later Jaina doctrine).71 This kind of renunciation is, however, not considered possible until a person has detached himself from the 'external possessions' - land, houses, silver, gold, etc. So the layman expresses his seriousness about aparigraha by setting limits to what he may own (i.e. he gives himself less to be attached to and, at the same time, engenders an attitude of non-attachment towards what he already has). 72 (It is interesting to note, however, that, even at this relatively late doctrinal stage, the renunciation of passion is approached, not via a direct confrontation with the internal state, but through the renunciation of activities
| 69 bẫhyānām gomahiyamanimultaphaladĩnăm cetanặcetanặnăm ābhyantarāņām ca rāgādinām upadhīnām samrakşaņārjanasamskārādilaksanāvyāvrttir mūrcchā - SS on TS 7:17. 70 tanmūlāh sarve doşāḥ ||
mamedam iti hi sati samkalde samraksanādayah samjāyante
tatra ca himsā 'vasyambhāvini |- SS on TS 7:17. Cf. J.L. Jaini (1920) on TS 7:17: ...'worldly objects are said to be Parigraha because they are the external causes of internal attachment'.
71 JPP p. 177; the aṇuvrata aparigraha is defined variously as 'non-possession', 'non-attachment', 'the delusion of possession', 'harbouring false notions of "this is mine'', etc. See discussion of Das. 8:37, p. 37ff., above, for the relation of parigraha and kasāya. Nokasāya, 'sentiments', are nine subsidiary passions, such as laughter, fear sexual cravings, etc. - see JPP p. 120 for a full list.
72 See JPP p. 177.

Page #89

Umāsvāti's Jainism ' 75 which are thought to generate it.)
As P.S. Jaini puts it:
By undertaking the aparigrahavrata, a Jaina layman systematically reduces his tendencies to fall into such passions; thus he protects his soul from increased karmic entanglement and lays the groundwork for complete nonattachment, the path of the mendicant. 73
How does this change in attitude towards the householder relate to the development of the kaşāya doctrine? The following hypothesis is offered.
In the early karma doctrine, only one train of events is necessary for karmic bondage: parigraha or lobha (including the other passions) causes himsā, and himsā causes bondage. (It should be noted that parigraha is by no means the only way in which himsā can be brought about, but it is seen as being the major threat to the monk, and the one most difficult to counteract.) In the developed doctrine, presented by Umāsvāti, for bondage to take place (i.e. for karma to attach itself to the jīva) there have to be two separate occurrences: 1) there has to be yoga ('activity') which causes an influx of karmic matter, and 2) there has to be kaşāya which causes that karmic matter to adhere to the jīva. Thus 'greed' (lobha or parigraha) - the passion subsuming all others in the earliest texts -, which was previously seen as being synonymous with a particular way of life (the householder's), becomes in the later doctrine (under the technical term 'kaṣāya') an internal process or attitude. From being the defining characteristic of a particular way or condition of life, it comes to denote anattitude towards life; the emphasis is shifted from the external to the internal, from the social to the individual. In effect, renunciation is partially internalised.
73 JPP p. 178.

Page #90

76 Harmless Souls
In this way, what was once an anti-householder doctrine becomes one which accommodates and compromises with the householder's way of life. In other words, parigraha and aparigraha have been redefined as attitudes, inner states; therefore it is possible, at least in theory, to retain possessions without necessarily incurring bondage. With the right attitude (non-attachment), possessions do not necessarily lead to himsā and bondage.
Nevertheless, Jaina monks and nuns, by definition nonhouseholders, continue to concentrate on behaviour: their attitude to the external world is significant predominantly in so far as it governs the way in which they behave in that world. Thus, in practice, for the ascetic, attitude remains a contributory rather than a necessary factor in liberation or bondage, in the same way that kaṣāya was in the earliest doctrine. (But as with kaṣāya, that of course makes it is no less demanding of the most careful attention.)
It can be seen from this that the later doctrine, which accommodates the householder, is never completely integrated with the earlier one. And as we have shown, there are instances in the Tattvärtha Sutra where the latter pushes through the surface of the former to bring about an apparent inconsistency. To take a relevant example, at Tattvärtha Sutra 9:6 śauca (purity) is named as one of the constituents or types of that 'dharma which, in turn, is one of the causes of samvara (the stoppage of karmic influx). Commenting on this, the Sarvārthasiddhi specifically defines sauca as 'complete freedom from greed'.74 As has been made clear above, greed (parigraha / lobha) was considered by the earliest surviving canonical sources as one of the two worst sins; moreover, it was considered to be the defining characteristic of the householder's way of life. It is greed which causes a person to undertake
prakarṣaprāptalobhan nivṛttiḥ saucam - SS on TS 9:6.
74

Page #91

Umāsvāti's Jainism 77 harming activity. Thus absence of greed means the absence of harming activity and of the potential for it. It is this condition which is considered to be a state of purity, and by definition it is only the monk or nun who experiences this greed-free state. In other words, in the earliest texts greed is seen as the polluting, himsā-initiating, bondage-causing vice. This connection is retained in the Tattvārtha Sūtra, where lobha is, by implication, synonymous with impurity, and freedom from greed is defined as sauca, although now the context has been internalised, in that the emphasis is placed on the stoppage of kaşāya, the attitude which is seen as instrumental in bondage. However, it is clear that originally the distinction between 'pure' and 'impure' was an existential distinction between the life of the monk and the life of the householder. And this earlier distinction resurfaces in the Sarvārthasiddhi when sauca is defined as 'freedom from greed'.
There may also be a suggestion here of a polemical statement aimed at Brahmanical ideas of what constitutes purity and dharma. That is to say, the Jains are reinterpreting sauca in a manner which shows that they are the truly pure, and it is the brahmans who are succumbing to greed, the antithesis of purity, just when they claim to be most pure. In both cases, impurity leads to further karmic bondage; the difference arises over how purity is defined. Brahman sacrificers cause himsā, therefore they are impure. Himsā (= impurity) is generated by greed, i.e., greed for results from the sacrifice. It is sacrifice, the characteristic mark of the śrauta brahman, which makes the brahman impure in the Jaina's eyes. And from the Jaina ascetic's point of view, the brahman is doubly impure, for in order to be qualified to sacrifice he has to be a householder.
In short, the behaviour of Jaina ascetics continues to be conditioned by the earliest beliefs about the conditions under which bondage takes place. (This is, of course,

Page #92

78 Harmless Souls progressively true of advanced lay behaviour, as it approaches, through the pratimās, etc., the ascetic ideal.)75 They remain, above all, concerned with the effects of physical activity. Thus, while from the ordinary householder's point of view there is a continuum between his aņuvratas and the mahāvratas of the monk, from the monk's point of view there is still an absolute distinction. This basic incompatibility of the two ways of life underlies Umāsvāti's attempt in the Tattvārtha Sūtra to reconcile them.
75 See JPP p. 186 for a full list if the eleven pratimās. Note especially: 8) ārambhatyāga-pratimā, the stage of abandonment of household activity, and 9) parigrahatyāga-pratimā, the stage of abandonment of acquisitiveness (by formally disposing of one's property).

Page #93

Conclusions
Commenting on the juxtaposition of the archaic and the classical in Jaina doctrine - on the tradition's incompletely worked-out philosophy and its tendency to fantastic proliferation - Frauwallner concludes that all this is the result of fundamental adherence to the doctrines proclaimed by Mahāvīra.1 Because the Jina is omniscient, his doctrines, once uttered, could not be changed or displaced, and that 'explains the many antique features which the system has preserved'.2 There was no room for consistent developments in thought, or for 'the erection of a uniformly compact doctrinal edifice'.3 In these circumstances, according to Frauwallner, it is not surprising that wherever the 'traditionally handed-down dogmatics showed a lacuna', fantasy was allowed to flourish without a check.4 In short, Jaina thought after Mahāvīra was paralysed by the need to preserve the often archaic content of Mahāvīra's teaching in more sophisticated religious and philosophical circumstances.
There is evidently some truth in this analysis, but as it stands it remains an inadequate explanation because Frauwallner has divorced Jaina beliefs and doctrines fro: practice; he is concerned with Jainism as a 'philosophy rather than as a religion. The two levels of doctrine regarding activity, the influx of karma and bondage, which are imposed upon each other and which fit so incompletely, are the result of two different historical processes - within Jainism and within Indian religion in general - which come together over a particular period. These two processes may, for short, be labelled 'Early Jainism' and 'Umāsvāti's
Frauwallner (1973) Vol. II, p. 213.
2 Ibid. 3 Ibid.
4 Ibid. p. 214.

Page #94

80 Harmless Souls Jainism'. By 'Early Jainism', is understood that exclusively ascetic, mendicant path to liberation which appeared at approximately the same time as other heterodox systems, notably Buddhism, partly as a reaction to Brahmanical religion, but, in Jainism's case, perhaps mostly as a refinement of an even earlier, archaic asceticism.5 'Umāsvāti's Jainism', on the other hand, belongs to a much wider social world, in the sense that it is an attempt to systematize, as far as possible, Jaina doctrine for the whole Jaina community, and perhaps most of all for a growing lay audience. That is to say, Umāsvāti is attempting to reconcile the social fact of an active lay following, and the need to preserve such a following, with a body of canonical texts, the oldest and most important components of which (containing perhaps the teachings of Mahāvīra himself) were directed specifically at ascetics who had renounced the householder's world precisely because, as the doctrines expounded in those texts make clear, there was no possibility whatsoever of obtaining liberation within it.
It is also significant that Umāsvāti is writing in Sanskrit. Thus he is not writing simply for the benefit of his own community, but also in order to dispute with outsiders, proponents of rival darśanas. In other words, for Jaina practice to be preserved and defended, its doctrinal superstructure has to be defended. Moreover, since Jaina renouncers keep moving, and are unlikely to know or learn Sanskrit, the very nature of Umāsvāti's enterprise suggests that it is concerned with problems in the wider society.
The whole history of Jaina doctrinal development is one of the struggle to prevent the clearly delineated but extremely demanding requirements of the ascetic's path coming into direct conflict with the life led by the Jaina layperson, when the practice of both ascetic and laity is recoded in doctrinal terms. It is precisely this conflict and the need to avoid it which leads, on the one hand, to the
*See Basham for some discussion of 'Jainism' before Mahāvira; also R. Williams 1966, pp. 2-6.

Page #95

Conclusions 81 proliferation of doctrines to do with karma and bondage, and, on the other, ensures that no complete systematization is possible.
As has been made clear, in the very earliest texts, where there is effectively no alternative to extreme asceticism other than a bad rebirth, there is no need for doctrine to be worked out in any great detail. From the ascetic point of view, which is the 'Jain' point of view at this stage (the earliest texts were composed by and for ascetics), the chance of a lay person attaining a heavenly rebirth is considered so remote as to be practically impossible. Proliferation of heavens and hells tends to occur only when a system becomes concerned with something more than simple liberation, i.e. when it acquires a lay doctrine. And only then do ideas like that of śubha yoga leading to punya karma become intelligible. (When the path to liberation is as demanding as the Jain one, ascetics too, of course, become interested in the higher samsāric worlds.)
Given the basic teachings of early Jainism on the multiplicity of jīvas, ahimsā, and the bondage of material karma, then the ascetic's path is theoretically very clear: he or she must give up all harming activity. Whether this in effect means all activity, or whether there is a distinction between activity of different kinds, is not clear from the earliest texts. It is probably taken for granted that in this context any action which harms jīvas is meant, and thus potentially, at the very least, any physical activity is harmful.
There is, therefore, a sense in which, beyond a few normative teachings of early Jainism, all doctrinal elaboration is aimed at a lay audience, both to justify their position as laity and to put them on the soteriological 'ladder'. This is reflected by the fact that doctrines concerning the internalization of significant action, which if carried to their logical conclusion might be thought to undermine, or even be fatal to ascetic practice, actually make no difference to the ways in which monks and nuns behave. They go on acting as though the overwhelmingly

Page #96

82 Harmless Souls important thing is physical activity, or the lack of it, in the external world: attitude is only important in so far as it leads to certain kinds of behaviour.
The doctrinal content of the earliest canonical literature is simply intended to reinforce ascetic practice. The problem for Umāsvāti is to reduce the incompatibility of a purely ascetic 'doctrine' with the householder's life, without juxtaposing them in such a way that they are seen to be openly opposed. In other words, he has to systematize, but he has to do so incompletely or imperfectly.
It is this necessary incompleteness which gives rise to much of the problematic and poorly fitting terminology in the Tattvārtha Sūtra. Two different historical layers, reflected in two different kinds of behaviour, are imperfectly systematized in the one work by a process of partial internalization, and therefore of gradation, of the path to liberation. This is most evident when Umāsvāti takes earlier 'lists', or concepts, and attempts to deposit them without change in new categories. It is not surprising, therefore, that later commentators have struggled without success to integrate then fully, and that in the process original meanings have been obscured or lost. The problematic nature of some Jaina terminology thus arises out of the need to revise earlier ideas in, and for, a more sophisticated and complex religious milieu, while retaining that canonical authority which is expressed in the ideal of ascetic practice. In other words, such terminological problems stem from the extreme asceticism of early Jaina practice and its incompatibility with lay practice: the two positions are so antithetical that they cannot both be preserved and at the same time fully incorporated within the same system.
The extent to which Umāsvāti internalizes earlier doctrine, and the effect of this process on both lay and ascetic practice, requires some further discussion. According to classical Jaina theory, it is possible for a lay person to ascend by stages to the threshold of total asceticism, and then pass through mendicancy to liberation.

Page #97

Conclusions 83 This is expressed theoretically by a soteriological gradation or ladder, whose rungs are the 14 gunasthāna, the stages through which an individual passes on the way to mokşa.6 Most, although not all, of the component parts of the gunasthāna doctrine are known to Umāsvāti, but they are not assembled into a śreņi or 'ladder' by him; while various karmic states are described individually, they are not placed in a hierarchy.7
According to the developed theory, the aspirant gradually eliminates the passions, and it is only finally, in the instant prior to his death, that yoga, the last cause of bondage, comes to an end. 8 Thus the problem of action - of how not to jeopardize one's ultimate liberation while at the same time acting to a greater or lesser extent in the world - is overcome by making 'activity' as such (i.e. unmotivated, unimpelled activity, even down to the beating of the heart) a barrier to liberation only on the final rungs of the ladder, which are usually taken as meaning the last moments of life. (There are more immediate barriers to be overcome before that stage is reached.) One of the reasons theoreticians after Umāsvāti are able to shift yoga up the ladder in this way is because the kaşāya doctrine makes the lower rungs the province of passions and their elimination. In other words, such a doctrine makes room for the theory of a graduated path of spiritual development, culminating in the elimination of all activity. This should be compared with the earliest, ascetic 'Jainism' - that of the Jaina monks - where the elimination of harming activity, the vow of total ahimsā, is regarded as the pre- or accompanying-condition of any spiritual development.
The stages of spiritual progress which lead the layman
6 For the full list, see JPP pp. 272-3.
7 See Ohira pp. 99-103. Also on the evolution of the guṇasthāna doctrine, see Dixit 1971, pp. 14-15. The doctrine as later developed is mentioned in the Bhāsya, which, as we have seen, is probably later than the TS and not by Umāsvāti (p. 46-47, above); the SS on TS 9.1 has a complete list.
8 See JPP p. 273.

Page #98

84 Harmless Souls into the sixth gunasthāna, which is the stage of taking the ascetic's vows (mahāvrata), are further subdivided into 11 pratimā.9 In other words, certain kinds of behaviour and attitude will enable the aspirant to make some progress towards liberation, but these have to be superseded or improved upon as he moves up to the next 'rung'.
The theoretical stress on internal discipline as the chief means of ascending the ladder clearly reflects the difficulty of achieving a completely inactive state short of death (i.e. short of the ideal goal of ritual death by fasting, sallekhanā). 10 It is true that to achieve final liberation the complete cessation of first external activity and then all activity has to be achieved, but only at the very end of life. For the laity, therefore, yoga has been effectively downgraded as a force in bondage - or, to be accurate, 'upgraded' so far that it is only to be reckoned with at the top of the soteriological ladder. It seems that they scarcely have to worry about it.
For the ascetic, however, the emphasis of the earliest doctrine on the centrality of physical action vis à vis bondage, keeps its force. The partially internalized doctrine, where what counts most is intention, is at a lower level soteriologically than the physical and material concerns of the monk. What distinguishes the monk most clearly from the lay person in 'classical' Jainism, in terms of daily practice, is still the strength of the monk's vow (vrata) concerning ahiņsā.11 And the authority for this mahāvrata of ahimsā is derived predominantly from the earliest canonical texts (the Ayāramga, etc.), i.e. it is the original doctrine concerning non-violence, aimed exclusively at renunciants.
It should be remembered, however, that the canonical texts, which reflect the monks' view of the world, and the Tattvārtha Sūtra, which (as I argue) reflects a wider view,
9 See JPP p. 186 for a full list; and R. Williams 1963, pp. 172-181. 10 On sallekhanā, see JPP pp. 227-223. 11 See JPP p. 241.

Page #99

Conclusions 85 co-exist. The latter does not replace the former. This reflects the way in which the two doctrines concerning bondage co-exist within the one religious system. There are two different authoritative references, depending upon whether one is a householder or a monk; but these only conflict when they are viewed together, theoretically; in practice, the perspective is always from within one or the other of these.
Nevertheless, even within the new soteriological gradation, it is only necessary to look at the vows (vrata - literally, 'restraints') which constitute even the lay pratimā, in order to see what is considered important in terms of practice, as opposed to theory. The emphasis of these is heavily ascetic, i.e. they involve above all the restriction of physical activity. In fact they constitute a progression towards the full asceticism of the monk or nun, carrying the vratin further and further away from ordinary householder existence.12 So whatever the theoretical stress on internal discipline or dispassion, in terms of actual conduct there is no remission of physical asceticism; rather such conduct is extended into the lay sphere. The ladder is lowered from above, not erected from below. P.S. Jaini, commenting on the Chedasūtras, writes that they 'provide valuable insight into the numerous restrictions imposed upon itself by the [monastic] community, mainly in order to preserve its integrity in the face of increasing dependence upon the laity'.13 I would add to this that another way to annul such a threat is for the ascetics to encourage the laity to become more like them. more ascetic. Whatever one's internal
12 See JPP p. 187 for a list, and ibid. pp. 157-187 for a discussion of how these relate to the pratimā. Jaini relies here upon R. William's study of mediaeval śrāvakācāra texts, Jaina Yoga (1963) OUP. See also TS 7:19ff. for householder vrata; these have been identified by R. Williams 1963, p. 2-3, as essentially Digambara vrata, whereas the 'autocommentary' is markedly Śvetāmbara in tone (ibid. p. 2, fn. 1), providing further evidence that the TS and its bhāşya are not by the same hand.
13 JPP pp. 63-64.

Page #100

86 Harmless Souls state, the only way of confirming it or expressing it, either to oneself or to others, is through external conduct. (I shall return to this idea in my discussion of Kundakunda.) 14
As P.S. Jaini remarks, 'even the clerics of many religions do not live so strict a life as these rules [of lay conduct] demand'.15 Consequently, as he goes on to say:
the partial vratas and the pratimās, while theoretically set down for all laymen, tend to constitute an ideal path followed only by a highly select few ... it is a rare individual who actually vows to accept the restraints or perform the holy activities described there. 16
It is clear from this situation - one which only develops fully in the post-Umāsvāti period - that Umāsvāti's internalization of discipline, through his kasāya doctrine, does not or is not allowed to influence ascetic behaviour. In so far as the monk or advanced lay person takes on internalized doctrines, he does so in addition to his original vrata, not instead of the latter or as a version of them.
A further consideration which may be mentioned here - again, I shall return to it in my discussion of Kundakunda - is the 'ritualisation' of ascetic conduct. By this I mean the idea that if one follows the prescribed action to the letter, the result in the Jaina case, liberation) is guaranteed. Correct thinking and correct feeling may be essential to the correctitude of a ritual, but the only way these can be monitored or expressed is through external behaviour. The īryāpatha-sāmparāyika division in the Viyāhapannatti apparently reflects at least a degree of such ritualisation. There the initial division between the two modes of āsrava / karma is not so much based upon the results of two modes of behaviour available to the same individual, as upon the institutional distinction between two modes of life,
14 See, for example, pp. 217-224, below. 15 JPP p. 188.
16 Ibid.

Page #101

Conclusions 87 incorporating two roles (that of the lay-person and that of the monk) which are mutually exclusive and which, in soteriological terms, inevitably have different outcomes.
Umāsvāti's introduction of a technical kaşāya doctine, linked to the īryāpatha-sāmparāyika distinction, would, however, seem at the very least to make room for a less mechanical and more attitude-dependent approach to ascetic conduct. For in theory, the kaṣāya doctrine bridges or blurs the institutional distinction between ascetic (iryāpatha) and lay or non-ascetic (sāmparāyika ) activity, since greater or lesser degrees of bondage now depend on the degree of passion accompanying or motivating an action, not on who performs it. Indeed, taken to its logical conclusion the kasāya doctrine would seem to undercut the rationale for ascetic behaviour altogether. (Clearly, internalisation can only go so far before it threatens the whole basis of a monastic system grounded in severe physical restraint, and with it any specific 'Jaina' identity.)17 But as we have remarked, it has no such effect, and it is probably not intended that it should have. What then is its purpose?
I suggest that the real importance of Umāsvāti's kaṣāya doctrine is for the ordinary lay person (i.e. the majority of the Jaina community), in that it provides a rationale for ordinary lay conduct. That is to say, once the principle that bondage is the ineluctable result of physical harm done to jīvas has been modified, and the principle that only actions motivated or actuated by passion are binding has been accepted, then it becomes possible for one to lead a good life, with good soteriological prospects, without necessarily abandoning one's position as a householder. Thus by controlling one's passions or attitude while performing a particular action, one can reduce or avoid the karmic effect associated with it. Or, to put it in the overlapping but not entirely integrated terminology, one can accrue punya as
17 See Kundakunda sections below, passim, for a full discussion of this.

Page #102

88 Harmless Souls opposed to pāpa karma. In other words, when they are allied to internal control, one's actions in the world are not necessarily counter-productive. A householder's life can now be viewed positively: there is such a thing as a (relatively) good action. This is particularly significant with regard to aparigraha: an attitude of non-possession while living in the world is not necessarily a contradiction in terms, providing the emotions associated with possession of objects are controlled. (And it is interesting to note that the Sarvārthasiddhi on Tattvārtha Sūtra 7:19 refers to a 'psychical home' (bhāvāgāram) - home-focused thoughts (caused by conduct-deluding karmas), the abandonment of which constitutes true 'homelessness'.)
As we have seen, the guṇasthāna and pratimā ladders formalize in theoretical terms the idea of gradation in spiritual - in practice, increasingly ascetic - progress. The householder is now at least on the same ladder as the ascetic. Umāsvāti's kasāya doctrine supports this in a tangential way; but, potentially, it is of most significance for those on the very first rungs of that ladder, the majority of the Jaina community. Through it, their affective subscription to Jainism's distinctive ethical code (preserved by the ascetics and exemplified in their behaviour) is not disqualified by their status as householders. In other words, their identity as 'Jains' is given theoretical (effectively doctrinal) validation. Some of this Jaina majority will, of course, make progress up the pratimā / guṇasthāna ladder, and as they do so - as their lives become more ascetic - the greater will be the importance of external conduct and the less relevant internalization on its own.
It is possible to conclude, therefore, that the doctrines found in the Tattvārtha Sūtra concerning the mechanism of bondage and the significance of yoga are aimed at the Jaina community as a whole. From the fact that Jaina ascetics continue to behave in the manner enjoined in the earliest canonical literature, it is clear that Umāsvāti's incomplete systematization is not so much evidence of a compromise between lay and ascetic life - indeed, the ascetic cannot

Page #103

Conclusions 89
compromise in this respect without abrogating the whole import of the canonical doctrine of ahimsa - as evidence of the need to construct a common doctrinal framework in which both ways of life are justified without nullifying each other by too close a contact. (The ascetic's authoritative reference remains, of course, the canon and/or the tradition which is embodied in ascetic practice.)
It may be that this process goes hand in hand with the need for the Jaina community to live in concord with Brahmanical society as a whole, while retaining its individuality. The reaction against Brahmanism evidenced in the canonical texts is a reaction of ascetics against specific Brahmanical practices. But for the laity to survive, there was an increasing need to over-code Jaina behaviour and doctrine to present a more Brahmanical front, while at the same time preserving the true purport of Jaina teachings. The beginning of this process may perhaps be seen in the development of the guṇasthāna theory, which, in terms of its function, seems analogous to the orthodox āśrama doctrine. The culmination comes with Jinasena's attempts to integrate the Hindu samskāras into Jaina lay practice in his Adipurāṇa (perhaps written to justify an already existing state of affairs).18 But it is in Umāsvāti's Tattvärtha Sutra that the first real attempt is made to present Jaina doctrine as an autonomous religious system which includes both monks and laity, rather than simply the teachings of a heterodox renouncer. It is here that one first has the sense of a community being addressed, rather than a collection of individual monks. And it is here too that Jainism can be seen consciously addressing outsiders. In other words, it now turns to face the wider Indian religious world which surrounds and threatens to infiltrate it.
With the works ascribed to Kundakunda we find Jainism somewhere near the centre of this terrain, subject to both internal and external pressures and developing social
18
See JPP p. 291ff., and R. Williams 1963, pp. 274-275.

Page #104

90 Harmless Souls and philosophical strategies for dealing with them. And it is to these that I now turn.

Page #105

PART III KUNDAKUNDA: THE PRAVACANASĀRA
4
Kundakunda: content and context
4.1 Kundakunda: primary sources and chronology
In this and the following part I shall examine two of the major Prākrit works ascribed to the Digambara ācārya Kundakunda, the Pravacanasāra and the Samayasāra.
Giving even an approximately accurate date to Kundakunda presents formidable problems. The revered and influential position he holds within the Digambara tradition only adds to the difficulties. Traditionally, two dates are ascribed to Kundakunda: the middle of the third century C.E. (fl. 243 C.E.), and the first half of the first century C.E., the latter being the more popular. Upadhye shows that the traditional evidence for these dates is drawn largely from much later commentators and is totally inadequate. Moreover, the idiosyncratic nature of some of Kundakunda's teaching in the context of the rest of the Jain tradition makes comparison with such texts as the Tattvārtha Sūtra unhelpful in this respect. And as we shall see, the nature of Kundakunda's texts is such - they are clearly compilations of older material held together by new philosophical and soteriological strategies - that it is difficult to remain confident that all or even any of them should be ascribed to a single author or redactor. Conversely, there are close enough thematic links between the Pravacanasāra and the Samayasāra to make it obvious that they originated in the same religious and philosophical milieu. They can, therefore, profitably be studied together.
The evidence for Kundakunda's date has been most
1 Upadhye 1935, p. xff.

Page #106

92 Harmless Souls extensively reviewed by Upadhye in his introduction to the Pravacanasāra.2 Here I shall confine myself to a brief summary of Upadhye's main arguments, and then offer some criticisms.
Upadhye remarks that Pūjyapāda, the earliest Digambara commentator on the Tattvārtha Sūtra, quotes (at Sarvārthasiddhi 2:10) five gāthās which are found in the same order in the Bārasa-Aņuvekkhā (25-19), a work ascribed to Kundakunda. Pūjyapāda, however, 'does not say as to from what source he is quoting'.3 Nevertheless, Upadhye thinks that their context and serial order indicate the genuineness of these quotations, i.e. Pūjyapāda is quoting from Kundakunda. If this is so then it would set a later limit to the age of Kundakunda since, according to Upadhye, Pūjyapāda lived earlier than the last quarter of the fifth century C.E.4 Upadhye then reduces this upper limit by reference to the Merkara (Mercara) copper plates of śaka 388 (466 C.E.) which mention a Kundakundānvaya, giving the names of at least six ācāryas of that lineage.5 This 'indicates that Kundakunda will have to be put at least a century, if not more, earlier than the date of the copperplates' (i.e. c.350 C.E.).6 Upadhye then assumes that the lineage of a saint does not begin immediately after his death, and so takes the date back a further 100 years. Thus his later limit is now c.250 C.E. He concludes, however, 'I am inclined to believe, after this long survey of the available material, that Kundakunda's age lies at the beginning of the Christian era' (his italics).8 In other words, he reverts to the traditional dating, albeit for
2 Ibid. pp. x-xxiv. 3 Ibid. p. xxi.
4 Bronkhorst p. 161, says that he lived 'not long after 455 A.D.'; see above, p. 46.
5See Upadhye pp. xix - xxii, where he refers to Epigraphia Carnatica I, Coorg Inscriptions No. 1., B.L. Rice, Madras, 1914.
6 Upadhye p. xix. 7 Ibid. p. xxii. 8 Ibid.

Page #107

Kundakunda: The Pravacanasāra 93
different reasons to those traditionally advanced.9
In support of this, Upadhye offers some linguistic evidence, namely, that the Prakrit dialect used in the Pravacanasāra, for instance, seems to be earlier than that of the Prakrit portions of the Natyaśāstra of Bharata (usually assigned to the beginning of the second century C.E., but, as Upadhye admits, the date is uncertain).10 He also claims that not a single Apabhramsa form is traceable in the Pravacanasāra, possibly indicating a period when Prakrits had not yet developed Apabhramsa traces.11 (He has already noted the strong influence of Sanskrit on the Prakrit dialect of the Pravacanasāra dubbed 'Jaina
Śauraseni' by Pischel: e.g. it retains intervocalic 'c' and even 'p' at times; desi words are also conspicuously absent.)12
I see the following problems with Upadhye's arguments. First, the ascription of the Bārasa-Anuvekkhā to Kundakunda can itself only be made on traditional grounds. Furthermore, as Upadhye himself admits, 13 the text may be a compilation of traditional gāthās, and nothing in the content of the verses quoted by Pujyapada indicates that they are necessarily non-traditional material.14 Consequently, we do not know that Pūjyapāda is quoting from Kundakunda. On these grounds, even the upper limit is uncertain.
Second, the Mercara copper plate inscription is now considered to be a forgery of the eighth or ninth century. Chatterjee describes these plates as 'definitely spurious'.15
9
Cf. Keith's 1936 review (pp. 528-9) of Upadhye's edition of the Pravacanasara. He concludes from Upadhye's own evidence that Kundakunda 'may be placed not later than the fourth century A.D.', but how much earlier than that is not clear.
10 Upadhye p. xxiii. 11 Ibid.
12 See ibid. pp. cxx-cxxi, cxxiv.
13 Ibid.
p. xl.
14 See SS on TS 2:10.
-

15 Chatterjee p.234, referring to Epigraphia Carnatica (revised ed.) 1972, Vol. I, Introduction pp. xf.

Page #108

94 Harmless Souls He goes on to argue that even if this plate were a genuine copy of an older record, by assigning twenty-five years to each of the six monks named, the earliest, Guņacandra, would not be before 325 C.E. So, on the basis of Pūjyapāda quoting from the Bārasa-Aņuvekkhā, Chatterjee assigns Kundakunda to the fourth century C.E. He remarks, however, that 'the anvaya of Kundakunda appears only in the records of South India, which were inscribed after 900 A.D.'. 16
Even if the Mercara copper plate were genuine, it only shows that there was an ācārya called Kundakunda, it does not indicate that the works we have that are ascribed to 'Kundakunda' are, either in toto or in part, necessarily by this particular teacher. Indeed, as Upadhye points out, 17 the only mention of Kundakunda's name in any of the works attributed to him is in the last gāthā of the BārasaAņuvekkhā, a verse which is not even found in some manuscripts. 18 Moreover, Amstacandra, his first extant commentator (c. tenth century C.E.) does not mention Kundakunda's name, and it is only with Jayasena (twelfth century C.E.) that a firm attribution is made.
Third, the linguistic evidence is highly approximate and largely depends upon unsure relative dating. It clearly needs much further research, but even then it would be unlikely to yield any chronological certainty.19 Schubring, however, in his article 'Kundakunda echt und unecht',20 compares the form and style of the Atthapāhuậa collection of texts (which are attributed to Kundakunda) with the form and style of other texts, particularly the Samayasāra. By
16 Chatterjee p. 325. 17 Upadhye p. ii. 18 See ibid. p. xl.
19 See F.W. Thomas, intro. to Faddegon's trans. (Kundakunda (3)] 1935, p. xix. Thomas puts Kundakunda in the third or fourth century C.E.; cf. P.S. Jaini, JPP p. 79, who favours the second or third century. See also Frauwallner 1973, p. 183 - fourth century C.E., Caillat 1987, p.508, dates Kundakunda to the first century C.E. without comment.
20 Schubring 1957, pp. 537-574.

Page #109

Kundakunda: The Pravacanasāra 95 doing so, he shows that the Atthapāhuda texts are much younger. He concludes, therefore, that the Atthapāhuda should no longer be considered a product of the classical period of Digambara literature and should not be classified as a work by Kundakunda.21 From their relative freedom from Apabhramsa forms, he takes works such as Samayasāra, Pravacanasāra, and Pañcāstikāya to belong to a genuine Kundakunda, whom, elsewhere, Schubring dates to the 2nd-3rd century C.E.22
None of this, however, brings us closer to a convincing date for the author(s) / redactor(s) of the Pravacanasāra and the Samayasāra. As Upadhye admits, 'we have to grope in darkness to settle the exact date of Kundakunda'.23
There is, however, some internal evidence as to the nature and chronology of these texts. E.H. Johnston in his study of early Sāmkhya notes in passing that:
Kundakunda's use of the terms pariņāma and paramāņu are more appropriate to a date in the neighbourhood of the third or fourth century A.D., and similarly in the Samayasāra .... 124, 127, and 356-361, he refers to the Sāmkhya doctrine of the connection between soul and prakrti in language that could hardly have been used at a much earlier date.24
My own research shows that the technical way in which the term samaya itself is used in the Samayasāra indicates a relatively late date (early fifth century or later) for that text.25 Furthermore, the way in which Kundakunda uses the two truths doctrine (vyavahāra-naya and niscaya-naya) seems much closer to Sankara's distinction than to the
21 Ibid. p. 574.
22 Schubring 1966, p. 36. Upadhye accepts the Pāhudas as genuine Kundakunda (see pp. xxvi-xxxvii), but Schubring convincingly dismisses this (1957, pp. 567-568).
23 Upadhye p. xix. 24 Johnston p. 14, fn. 1. 25 See below, p. 233ff.

Page #110

96 Harmless Souls Buddhist one. This alone might make one wonder how early parts of the Samayasāra can be dated. And a close examination reveals that this two truths doctrine is utilised in the Samayasāra in two different and ultimately conflicting ways - a clear enough indication that we are dealing with a composite text. In other words, the text as we have it has been subject to substantial modification and addition, probably as a result of non-Jaina philosophical influences.
This leaves as open questions the identity and date of the 'original' Kundakunda. But even if it were possible to answer these, it is unlikely that our understanding of the texts as we have them now would be significantly advanced. Moreover, my purpose here is to chart a particular development within Jaina thought and religion, and to define its practical limits. And for that project a cursory inspection of the Pravacanasāra and the Samayasāra shows that they are related within the same broad tradition, and so may be fruitfully examined in tandem regardless of specific authorship. Nevertheless, within that context, it is important to bear in mind the above problems and to assess each work, and even to some extent each gāthā, individually. As Upadhye admits, 'the compilatory character of Kundakunda's works nullifies the criterion whether a gāthā fits a particular context or not'. 26 He goes on to say that the available manuscripts are all accompanied by various commentaries and so are already 'under the bias' of particular commentators.27 In this respect, I shall refer in particular to the first extant commentaries on the Pravacanasāra (the Tattvadīpikā) and the Samayasāra (the Atmakhyāti), both by Amrtacandra, who is tentatively dated by Upadhye to the end of the tenth century C.E.28 I shall also occasionally refer to Jayasena's
26 Upadhye p. 1.
27 Ibid. p. li.
28 Ibid. pp. c-ci. Cf. R. Williams 1963, p. 24, and F.W. Thomas, intro, to Faddegon [Kundakunda (3)], p. xxiv, for other possibilities.

Page #111

Kundakunda: The Pravacanasāra 97 commentaries, which were probably written in the second half of the twelfth century C.E.29 There is therefore a considerable time-lag between the texts and their first commentaries. This may not, however, be as great as has usually been thought.
In what follows, I shall for convenience refer to both the Pravacanasāra and the Samayasāra as the works of Kundakunda, while keeping the above reservations in mind.
4.2 Upayoga
At the heart of Kundakunda's soteriology in the Pravacanasāra is the doctrine of upayoga (Pk. uvaoga). I shall begin this section, therefore, with a brief examination of the way in which this term is used in texts prior to, or approximately contemporary with, those attributed to Kundakunda. I shall then look in detail at the ways in which the meaning of the term was modified and developed by Kundakunda in the Pravacanasāra.
i) Upayoga before Kundakunda Among the earliest surviving texts there appears to be no direct reference to upayoga in the Āyāramga, the Sūyagadamga, or the Dasaveyāliya Suttas. The term does appear, however, at Uttarajjhayaņa XXVIII.10, where it is said that the characteristic of the jīva is manifestation (or application) through (or with) 'knowledge, perception, happiness and suffering'. 30 It is not clear whether upayoga, which can have the meanings 'application', 'manifestation',
29 See Upadhye p. liv. 30 jivo uvaogalakkhano
ņāņeņam damsaņeņam ca suhena ya duhena ya || Utt. XXVIII.10|| Jacobi translates this as 'The characteristic ... of the soul [is] the realisation (upayoga) of knowledge, faith, happiness and misery' - 1895, p. 153. For the meaning of damsaņa (darśana), see p. 98, below. P.S. Jaini (JPP p. 97) defines darśana as 'insight ... into the nature of reality (along with faith in this view)'. Cf. Pañc. 115, quoted p. 187, below.

Page #112

98 Harmless Souls
or 'employment', should be taken here in its full technical sense of 'application' or 'manifestation of consciousness'. The passage could mean that ṇāṇa, damsaṇa, etc., are particular types of upayoga, or that upayoga is one of a number of characteristics of the jiva; but in either case it need not mean 'consciousness' as such, although the assumption among commentators and translators is that it does.
In the following verse (Uttarajjhayaṇa XXVIII.11), which has a different classificatory system, it is clear that uvaoga (upayoga) is one among a number of characteristics of the jiva, and not yet the defining characteristic:
The characteristics of the soul are knowledge, faith, conduct, austerities, energy and application (uvaoga).31
This verse (11) seems to be a mixture of the prescriptive and the descriptive. It is also worth noting that damsana (darśana) can mean either 'faith' or 'perception', and although it is not clear which sense is being applied here, it is probable, since they imply each other, that it is being used in a non-exclusive way. Nevertheless, between them these two verses do contain in embryo all the elements of the later upayoga doctrine, although they are not yet or not explicitly arranged in a causal hierarchy. That is to say, they contain upayoga, jñāna and darśana, as well as subha and aśubha (assuming that sukha and duḥkha result from punya and papa which, in the later doctrine of Kundakunda, are the products of subha- and asubhaupayoga respectively). Thus these verses clearly originate from a period before the upayoga doctrine was fully developed, despite the fact that as a whole Chapter XXVIII was a relatively late addition to the body of the
-
31 ṇāṇam ca damsanam ceva carittam ca tavo taha |
viriyam uvaogo ya eyam jivassa lakkhaṇam || Utt. XXVIII.11 || Jacobi (ibid.) translates uvaoga here as 'realization (of its developments)'.

Page #113

Kundakunda: The Pravacanasāra 99 Uttarajjhayana. This is borne out by the fact that the formal division of upayoga into jñāna and darśana has not yet been made here, although, as we shall see, that particular classification can be traced back at least as far as the Pannavaņā Sutta.32
References to upayoga in the Viyāhapannatti (Bhagavati) take us much closer to the classical doctrine. At Viyāhapannatti II.10a (147b) we read that the guņa (essential property) of the atthikāya (fundamental entity) jiva makes possible uvaoga ('the spiritual function').33 And at II.10c (149a) it is asserted that the 'characteristic (lakkhaņa) of soul is the spiritual function (uvaoga)'. The soul by its own nature (āya-bhāveņam) possesses will (viriya) 'which enables it to apply this spiritual function in the infinite number of possibilities (pajjava) of cognition - viz. in the domains of the five knowledges, the three nonknowledges and the three visions (damsaņa) - thus revealing the true nature of soul (jīva-bhāva).34
At Viyāhapannatti XVIII.10e (760a) the brahman Somila asks Mahāvīra whether he is 'one or two ... imperishable (akkhaya), immutable (avvaya) and stationary (avatthiya) or has he different forms in past, present and future (aņega-bhūya-bhāva-bhaviya)?' Mahāvīra replies that:
He is all of these, since from the point of view of [the] essence [of his soul, Abhay.]35 (davv'atthayāe) he is one, from the point of view of knowledge and vision (nāņa-damsan'atthayāe) he is two; as to paesas (space-points] he is imperishable, immutable and stationary, but as to uvaoga he has different forms in past, present
32 See Schubring 1962, para.82; and below. 33 Deleu's 1970 trans. References are to his critical analysis.
34 Deleu II.10c. Cf. XII.4.4a (608a): 'The characteristic of the fundamental entity (atthi-kāya) soul (jiva) is the spiritual function (uvaoga-lakkh ...) which reveals itself in the different knowledges etc., ref to II. 10c.' For a description of these five knowledges, etc., see discussion of Pañc. 41, p. 101, below.
35 I.e. according to Abhayadeva's Vịtti on Viy.

Page #114

100 Harmless Souls
and future. 36
Here it is interesting to note that the form of uvaoga, as 'spiritual function', is something that changes; moreover, it is essentially something that the soul changes itself.37
Viyāhapannatti also uses the terms anāgāróvautta and sāgāróvautta to designate two types of uvaoga. Deleu, following Schubring's translation of sāgāra- and aņāgārauvaoga,38 renders these as 'faculty of abstract or indistinct imagination' and 'faculty of concrete or distinct imagination', respectively.39 At Viyahapannatti XXV.6.17 (899b) all classes of niyantha (monks) are said to have 'the formally distinct or the formally indistinct imagination (are sāgārôvautta or aņāgārôvautta)'.40 Uvautta is the Prākrit form of the Sanskrit upayukta, meaning 'employed' or 'applicable to'; similarly, upayoga, as we have seen, can have the meaning of 'application', 'manifestation', or 'employment' (of consciousness), and these are the translations I shall prefer.41 Deleu's use of 'imagination' (see above) to translate the term seems eccentric. But at Viyāhapannatti V.4 (221b) he renders uvautta as 'attentive'; and at XVIII.8.3 (755a) he gives sāgāra as 'formally distinct' and anāgāra as 'formally indistinct'. The Prākrit term āgāra corresponds to the Sanskrit ākāra, 'form' or
36 Deleu 1970.
37 Note also that jñāna and darśana share the same context as upayoga here, but they are not linked as explicitly as in the TS, for instance. See below.
38 See Schubring 1962, para. 82. 39 See, for instance, Viy. VI.3.5 (257b), and XIX.8 (770b).
40 Deleu 1970. If uvautta corresponds to uvaoga then this can hardly be true of monks alone, since uvaoga is the characteristic of all jīvas; it is, however, possible to use upayoga in the more limited sense of 'understanding'- see p. 102, below.
41 Frauwallner (1953, pp. 258, 287) translates upayoga by Betätigung, 'work' or 'activity'; but perhaps 'activation' / 'actuation' - and so 'awareness' (P.S. Jaini 1980, p. 223) and 'active consciousness' (Tatia pp. 55-56) would be better.

Page #115

Kundakunda: The Pravacanasāra 101 'figure'; thus, taking uvautta as 'attentive', sāgārôvautta would be 'attentive to something with form', and anāgārôvautta 'attentive to something without form' (i.e. determinate or indeterminate cognition / manifestation of consciousness), corresponding to the two kinds of upayoga, jñāna and darśana ('knowing' and 'perceiving').42
P.S. Jaini, following the Sarvārthasiddhi, has summarised the classical upayoga doctrine concisely.43 He states that the jīva has three main qualities (guna): caitanya, sukha and virya (consciousness, bliss and energy). Of these, consciousness is central, the distinguishing characteristic of the soul. Through the operation of this quality the soul can be the knower (pramāts), 'that which illuminates both objects and itself. Upayoga is 'application of consciousness' (which Jaini refers to as 'cognition').44 It is twofold, consisting of darśana (perception, first contact, or 'pure apprehension') and jñāna (comprehension of the details of what has been perceived).45
If we are looking for the origins of the technical use of upayoga to mean 'application of consciousness', it is worth noting that at Pañcāstikāya 41*1(42) and 41*2(43), for instance, upayoga is used in a somewhat different and apparently less embracing sense as a component part of matijñāna and śrutajñāna (the first two types of jñāna in the list of 5[8]), with the meaning of 'understanding of
42 For these see discussion of TS. 2:9 below. 43 JPP p. 104.
44 Cf. P.S. Jaini 1980, p. 223, where Jaini translates upayoga as 'awareness'. The purest upayoga (i.e. kevalajñāna) is connected with the siddha, and the impurest with the submicroscopic nigoda. Thus the degree to which one's upayoga is obscured by impurity exactly reflects one's place in the karmic hierarchy (ibid.). Also see Tatia, who translates upayoga as 'active consciousness' (pp. 55-6 and p. 55, n.5).
45 For controversies over whether these are two distinct gunas of the soul or simply aspects of the one, see Tatia, pp. 70-80. And for further details of the standard use of uvaoga / upayoga, see Schubring 1962, paras. 71 and 82.

For Private & Personal. Use Only

Page #116

102 Harmless Souls things' (i.e. 'application of knowledge'). Tatia comments that upayoga here means 'active consciousness', as opposed to labdhi, 'dormant consciousness':46 'soul is called upayukta v upayogavān when it is actually engaged in knowing something. Mere capacity for knowledge without actual knowledge is labdhi'.47 Although not relevant to the present enquiry, further research on this particular distinction might well shed light on the beginnings of the technical use of upayoga.
Turning to the Tattvārtha Sūtra, we find that upayoga is again defined as the laksaņa, the distinguishing characteristic of the jīva. So, in referring to the jiva, Tattvārtha Sūtra. 2.8 reads (in its entirety):
Application of consciousness is the distinguishing characteristic (of the sou!).45
In the next verse (TS. 2.9),49 application of consciousness (upayoga) is said to be of two kinds (subdivided into eight and four kinds respectively). These two categories, according to the Bhāsya and the Sarvärthasiddhi, are jñāna (knowledge) and darśana (perception); that is to say, application of consciousness with and without 'form'. 50
Enlarging on this, the Sarvārthasiddhi remarks that:
Apprehension of the mere object (the universal) is perception, and awareness of the particulars is knowledge. These occur in
46 Tatia p. 55. 47 Ibid. p. 56. 48 Upayogo lakṣaṇam [TS 2:8). 49 Schubring 1962, para. 82, traces this back to Pannavaņā 29. 50 upayogo dvividhaḥ sākāro anākāraś ca jñānopayogo
darśanopayogaścety arthah (Bhāsya on 2:9). Cf. use in Viy. above. Note also the use of sākāra and anākāra at Pravacanasāra 2:102 to denote, respectively, the roles of the ascetic and the layperson. The two usages are clearly connected, in that while jñāna can be seen to be more characteristic of the ascetic, darśana is relatively predominant in the lay person.

Page #117

Kundakunda: The Pravacanasāra 103 succession (darśana and then jñāna) in ordinary mortals (nonomniscients), but simultaneously in those who have annihilated karmas. 51
This accounts, in part, for the kevalin's omniscience. It also makes it clear that upayoga, when unobstructed by karmas, is instantaneous knowledge and perception (i.e. pure consciousness), the characteristic nature of all jīvas attained in kevalajñāna.
At Tattvārtha Sūtra. 2.8 we have seen that upayoga is a manifestation of what it is that differentiates the jīva from the non-jiva (ajīva) - its 'jivaness' or svatattva. In sūtra 2.1, however, Umāsvāti has described the jīva's inherent nature (svatattva) in terms of bhāvas of five kinds, arising from the four types or conditions of karma plus the jiva's natural or inherent bhāva (pāriņāmika bhāva), independent of karmas. According to this sūtra, the five kinds of bhāvas ('dispositions' or 'states of the soul') are:52
1) aupaśamika - arising from subsidence of karmas (of the
deluding kind), 2) kṣāyika - arising from destruction of karmas (of the four
ghātiya, or destructive kinds), 3) miśra (ksāvopaśamika) - arising from the destruction- cum
subsidence of karmas (of the destructive kind), 4) audayika - arising from the rise of karmas (the fruition of
karmas), 5) pariņāmika - that which undergoes modification, i.e. the
inherent nature / capacity of the jīva (independent of karmas).
We have already seen that at Viyāhapannatti II.10c the
51 Trans. by S.A. Jain p. 56. of:
sākāram jñānam anākāram darśanam iti / tac chadmastheșu krameņa vartate /
nirāvaraņeșu yugapat. 52 Following the SS on TS 2:1, which reads:
aupaśamikaksāyikau bhāvau miśraś ca jīvasya svatattvam audayikapāriņāmikau ca.

Page #118

104 Harmless Souls jīva-bhāva or 'true nature of soul' is said to be revealed through upayoga, which in turn applies itself through the five jñānas, etc. What then is the connection between the five bhāvas (i.e. bhāva used in a technical sense) and upayoga?
Perhaps the clearest line to take is that followed by P.S. Jaini in The Jaina Path of Purification,53 where he points out that in the standard Jaina doctrine of anekānta (manifold aspects) an existent (sat) is composed of three aspects: substance (dravya), quality (guna) and mode (paryāya) - a substance being a substratum (āśraya) for manifold guņas which, while free from qualities of their own, continuously undergo modifications (pariņāma) while acquiring new modes (paryāya or bhāva) and losing old ones.54 The point to note is that bhāva can be used as a synonym for paryāya. Applying this to the present case, we can see that consciousness (cetanā / caitanya) is a guna of the dravya jīva; upayoga is that consciousness manifested or applied in jñāna and darśana; the particular forms of jñāna (such as matijñāna, kevalajñāna, etc. (see below]) are thus paryāyas or bhāvas of the guņa jñāna (which is really an aspect of the guņa cetanā). So the various kinds of jñāna , for example, are bhāvas or 'modifications' of the jīva, reflecting the karmic condition of the jīva at that time - which itself was brought about by previous bhāvas. For this reason, bhāva is sometimes translated as 'thought-activity' or 'psychological disposition', but for a Western reader this could be misleading unless it is clearly understood that bhāva, as a modification or particular form of upayoga, is no more attributable to manas (mind) than is upayoga itself. (Tattvārtha Sūtra 2.11 is specific that there are two kinds of jīvas, those with minds and those without; whereas upayoga characterises all jīvas.) 'Manifestation of consciousness' and 'thought activity' should therefore be
53 Op. cit. pp. 90-91. 54 See TS. 5:29, 30, 38.

Page #119

: Kundakunda: The Pravacanasāra 105 taken in the sense of having consciousness (cetanā) as their base (āśraya). And as Kundakunda explains in his most orthodox work, the Pañcāstikāya [38, 39), cetanā is experienced by the three kinds of jivas in different ways: one kind of jīva (in fixed organisms and plants) simply experiences the fruits of karmas (kammāņam phalam ekko), another kind (embodied beings, etc. - ekko kajjam) experiences 'conative activity' as well, and another (the kevalin who is free from all physical and organic conditions) has 'pure and perfect knowledge' (nāņam adha ekko).55
In short, bhāvas are modifications or particular forms of upayoga. Thus by comparing the list of jñānas given in the Tattvārtha Sūtra with the five bhāvas (listed at Tattvārtha Sūtra 2.1) we can see the way in which the former are all essentially modes (bhāvas) of the jñāna component of upayoga.
The eight kinds of knowledge (sometimes divided into five right and three wrong) are as follows:
1) matijñāna - sensory 2) śrutajñāna - scriptural 3) avadhijñāna - clairvoyance 4) manah paryayajñāna - telepathy 5) kevalajñāna - omniscience 6) matyajñāna - wrong sensory 7) Śrutyajñāna - wrong scriptural 8) vibhangajñāna - wrong clairvoyance.
Corresponding to the other component of upayoga, four kinds of darśana are usually listed with the jñānas:
1) cakşurdarśana - perception through the eyes 2) acakşurdarśana - perception through the senses other than the
eyes
55 Following Chakravartinayanar (Kundakunda (2)].

Page #120

106 Harmless Souls
3) avadhidarśana - clairvoyant perception 4) kevaladarśana - omniscient perception. 56
This list sometimes occurs without either kevalajñāna or kevaladarśana;57 that is because these two are not strictly speaking separate types of jñāna and darśana (upayoga) but the substrata of all the others - pure jñāna and darśana without the impediment of karmas. Similarly, pariņāmika bhāva is not a bhāva or modification of the jīva as such, but the natural state of the jīva when it is free from karmic modification (i.e. from bhāvas ).58 To put this in terms of the classical karman doctrine, 59 bhāvas are states produced in the jīva by karmas. Thus five bhāvas are possible in the jīva 'which can manifest themselves simultaneously in a greater or smaller number'. 60 These basic five subdivide into a total of fifty-three possible states of the jiva, among which are the different kinds of upayoga (jñāna and darśana) and thus the eight jñānas. The amount or fraction of upayoga - which, it should be remembered, is the necessary characteristic of a jīva and so cannot be totally absent - is different in different beings, depending upon the degree to which their inherent kevalajñāna (or pariņāmikabhāva) is obscured by material karman. In other words, the five (fifty-three) bhāvas list all the
56 See TS 1 for a detailed discussion of the kinds of jñāna; also see the SS on TS 2:9, and Pañc. 41 which repeats the list exactly. .: 57 See, for instance, TS 2:5.
58 In terms of Kundakunda's doctrine (see below), true bhāvas (14) must be modifications of either śubha or aśubha upayoga; pariņāmika is free from karmas and thus identical to kevala- and darśana-jñāna, which comprise the state of suddha-upayoga. As explained on p.103, pariņāmika refers to the jiva's natural or inherent bhāva, independent of karman. Or to put it the other way round, being the natural condition of the jiva, it is that state which is the substratum of, or has the potential to undergo, modification under the influence of karman; but for that very reason, it is in itself free from such modification.
59 Schematised by Glasenapp p. 40ff. 60 Glasenapp p. 40.

Page #121

Kundakunda: The Pravacanasāra 107
possible states of the bound soul (i.e. its karmic states). Distributed among these states are the jñānas and darśanas which make up upayoga; that is to say, since the various jñānas and darśanas (apart from kevalajñāna and kevaladarśana) are essentially bhāvas, they are ultimately products of karmic bondage.
Before leaving the subject of bhāvas, it is interesting to note that Kundakunda at his most orthodox (in the Pañcāstikāya) equates, in terms of their function, bhava with kaṣāya, and that bhava here plays the role that is taken by upayoga in the Pravacanasara.61 So Pañcastikāya 147 (Upadhye's ed.; 154, SBJ edition) reads:
Whatever arisen state (bhāva) the impassioned self creates, auspicious or inauspicious, that state becomes the bond through that various material karman.
And the following verse [Pañcāstikāya 148 (155)] continues:
Physical combination (of karmic matter with the jiva) is caused by activity (yoga). Activity is of mind, speech and body. Bondage is caused by bhāva(s); bhāva(s) consist(s) of pleasure and attachment, aversion and delusion.
In other words, it is yoga which causes influx of material karman, and subha and aśubha bhāvas, consisting of passions, which are instrumental in that karman adhering to the jiva. This is simply the standard, two stage kaṣāya doctrine of the Tattvärtha Sutra, with the two components sometimes referred to as dravyabandha and bhāvabandha (the material and efficient causes of bondage). Thus, when upayoga is substituted for its equivalent, bhāva, it is clear that the upayoga doctrine performs precisely the same function in Kundakunda's soteriology as the kaṣāya doctrine in Umāsvāti's. This remains true while upayoga is
61 For bhāva in the Samayasara, see p. 267ff., below.

Page #122

108 Harmless Souls characterised merely as śubha and aśubha, i.e. while it refers exclusively to samsāric manifestations of consciousness. However, as will become clear, when the emphasis is switched from samsāric to nirvāṇic, from the impure to a pure state of consciousness, and the latter is viewed as a positive state (as opposed simply to the absence of samsāric states), Kundakunda's upayoga doctrine comes to provide a radically altered perspective on Jaina theory and practice - a possibility latent in the kaşāya doctrine but not activated.
It is possible in this way, and with hindsight, to trace the development of the upayoga doctrine (or at least the stages in which it is presented) in Kundakunda's works. For purposes of simplification, we can point to the passages just quoted as examples of an 'earlier' strand of the doctrine, 62 before a suddha / asuddha distinction is introduced. Nevertheless, the material for the 'later doctrine' is already in place. The eight jñānas and four darśanas (the component parts of upayoga) listed above (pp.105-6) can be divided, from a soteriological perspective, into more (6-8] (aśubha) or less [1-4] (śubha) 'negative' states (darśana [1-3] would be 'neutral'), counterbalanced by a 'positive' state (kevalajñāna and kevaladarśana). As remarked above, the latter is essentially freedom from karmic impediments, so it is not, strictly speaking, a manifestation or application of consciousness at all, rather it is the essence of the jiva unbound. Only śubha and aśubha upayogas are upayogas as such, since only they are the product of bhāvas, i.e. of karma-controlled states. The fact, already mentioned, that kevalajñāna and kevaladarśana are sometimes missing from the list of types of upayoga63 probably reflects the realisation of this, as well as giving a strong indication that this doctrine was modified over a period of time and only
62 Cf. my comments below on Pravacanasāra 2:63. 63 See, for example, TS 2:5.

.

Page #123

Kundakunda: The Pravacanasāra 109 reached its 'final' form with Kundakunda.64
The consequences of seeing kevalajñāna and kevaladarśana combined - the latter being subsumed in the former - as a form of (pure) consciousness, as Kundakunda does in the Pravacanasāra, are extensive, as I shall make clear. Here I shall restrict myself to the comment that, if it is the quality of consciousness that is instrumental in bondage and freedom (and the role of yoga [activity] has, via the two tier system of the Tattvārtha Sūtra and its precursors, become largely irrelevant to the actual mechanism of bondage), then it is only a relatively small step to saying that material karman's association with the jīva (i.e. bondage) is 'unreal'. For if the true nature of the soul is pure consciousness, then how can this be touched in reality by karman, which is material? And as will be seen, Kundakunda comes to assert that it is only from the vyavahāra-naya that the jīva is characterised by bhāvas, or śubha and aśubha upayoga (i.e. bondage through karmic matter); from the niscaya view, which in the Samayasāra, at least, he takes to be the 'real' view, the jīva has no connection with these. Soteriology thus becomes a matter of knowing and realising the true nature of the self, by means of jñāna and meditation. Tapas, on the other hand, has.- at least in theory - been down-graded. 65 Emphasis is switched from an obsession with the minutiae of the karma theory, where potentially every action has soteriological repercussions and interaction with the material world is crucial, to ways of achieving soteriological autonomy and
64 Three stages are suggested: 1) upayoga is used in the strict sense of 'manifestation of
consciousness', so the state of the kevalin is not included among the
jñānas and darśanas, the component parts of upayoga, 2) kevalajñāna and kevaladarśana are added to the list of upayogas, 3) kevalajñāna = suddhopayoga = the self, consciousness in its
'original', pure state, which is not a manifestation or application at
all.
65 For a general discussion of these strands in Indian religions, see Gombrich 1988, p. 44f.

Page #124

110 Harmless Souls by-passing the karmic world altogether through the realisation of its irrelevance, or even its unreality, in terms of the essential self. In the light of orthodox Jaina doctrine such a position - although perhaps logical given the premises on which the doctrine is founded - is, to say the least, startling, and will require considerable comment once the evidence has been considered.
To return to the present argument, we have seen that upayoga is the term used to designate 'application of consciousness', and that this is considered the defining characteristic of the jīva. However, so far upayoga has not been - and is not in standard expositions of Jaina doctrine which follow the Tattvārtha Sūtra (including most secondary sources in Western languages) - directly designated as the instrument by which the soul causes itself to be bound. Given that Umāsvāti in the Tattvārtha Sūtra identifies kaṣāyas (passions) as the efficient cause of bondage through karma, and that kaşāyas need some kind of initial consciousness to engender them, then it might be inferred from the doctrines of the Tattvārtha Sūtra that ultimately upayoga is 'responsible for bondage.66 This is still some way from saying that it is the quality of consciousness alone which is directly instrumental on every count in binding and freeing individual jivas. This in turn is equivalent to saying that the soul is totally responsible for and in control of its own bondage or freedom, given that upayoga is the characteristic of jīvas. To say so is not to maintain that particular states of consciousness cannot be, and are not, accounted for in terms of greater or lesser degrees of karmic obstruction, but that the cause of that obstruction comes to be seen in terms of the new central 'metaphor', i.e. consciousness itself, rather than in terms of particular physical activities and their consequences. And when pure consciousness is identified with an 'original',
66 Umāsvāti himself does not link the two, although the connection is evident from, for example, the way in which bhāva is used in the same context in the Pañcāstikāya (see above).

Page #125

Kundakunda: The Pravacanasāra 111. pure self, this inevitably comes to undermine a theory of bondage based upon material karma. In other words, when ignorance (ajñāna / avidyā) of the true nature of the self becomes the overriding factor in bondage, material karman itself starts to lose reality.
ii) Upayoga according to the Pravacanasāra Although what I shall call Kundakunda's 'upayoga doctrine' is used repeatedly in the Pravacanasāra to explain the mechanics of bondage and liberation, there is no one group of gāthās in which it is systematically explained or justified in philosophical or doctrinal terms. On the contrary, it is presented as though it were a commonly accepted doctrine in need of little direct explanation. In this it demonstrates its compatibility with the nature of the text in which it occurs, since the latter has more the appearance of a mosaic of 'traditional', or earlier material, arranged on a roughly thematic basis, rather than something composed as a unity. However, as far as I know, the upayoga doctrine does not appear in this form in any recorded source prior to Kundakunda. Indeed, commentators frequently remark upon the peculiarity, or uniqueness of Kundakunda in this respect. 67 For all hermeneutic purposes, therefore, he must be taken as the originator of this particular form of the upayoga doctrine.
The purpose of this compilation, at least in Books 1 and 2, is to instruct advanced mendicants (śramaņas) in the discipline which leads to self-knowledge and final liberation. Book 3, insofar as it is directed at all, seems to have a different audience in mind, namely, those just setting out on the śramaņas' path and advanced lay people. As I shall show, taken in the context of the upayoga doctrine of the first two books, there is, in terms of soteriology, something concessionary, and even
67 See, for instance, Tatia, p. 74.

Page #126

112 Harmless Souls contradictory, about Book 3; and this presents valuable evidence about the ways and extent to which the 'worldly' concerns of the laity were accommodated in the pattern of a largely 'internalised' ascetic soteriology.
Given the unsystematic nature of the Pravacanasāra, it is useful at the outset to give a schematic, if somewhat simplified version of the doctrine I shall go on to examine in greater detail in context. According to Kundakunda, upayoga, composed of jñāna and darśana, is the 'self of the self, the ātman of the ātman. It appears in two forms, as śuddha cr pure upayoga, and as aśuddha or impure upayoga. Ąśuddhopayoga is further divided into śubha (auspicious) and aśubha (inauspicious) upayoga. Suddhopayoga is the characteristic of liberated and liberating states of consciousness; aśuddhopayoga is the characteristic of bound and binding states of consciousness. Śuddhopayoga, the soteriological ideal, is the condition achieved or aspired to by the most advanced ascetics (i.e. the most practised in terms of inner discipline). Aśuddhopayoga is the condition of the majority (i.e. of ascetics short of the ideal, and of the laity). There is, however, a gradation in the quality of bondage in this latter condition, so that ascetics and advanced lay people have, or should aspire to bring about through particular kinds of behaviour, upayoga that is more śubha than aśubha. (At least, according to some gāthās they should aspire to do so; as we shall see, it depends upon which audience is being addressed.)
Upayoga is thus both instrumental in and typical of certain states of consciousness which are directly linked to both soteriological status and social role. Most importantly, it is seen to be directly and solely instrumental in bondage, something which differentiates its appearance in Kundakunda's works significantly from its use elsewhere (outlined above). In schematic form the doctrine may be

Page #127

113
Kundakunda: The Pravacanasāra presented as follows:
Upayoga (The 'self of the ātman = jñāna and darśana)
Liberated and liberating (i.e. pure) consciousness
Bound and binding (i.e. impure) consciousness
śuddhopayoga
aśuddhopayoga
śubha
aśubha
(The condition of the most advanced ascetics and the natural state of the self)
(The condition of less advanced ascetics and laity)
Bearing in mind Upadhye's warning that upayoga is 'a very mobile term, whose shade of meaning slightly changes according to the context',68 I shall deal first with the general features of upayoga as described in the Pravacanasāra, and then, in the next chapter, with its specific implications for the mechanisms of bondage and liberation.
At Pravacanasāra 2.35, the substance jīva (davvam jīvam) is described as cedanovajogamao, i.e. as consisting of the manifestation of consciousness;69 and according to Amstacandra's commentary on the Pravacanasāra, the Tattvadīpikā, this is the distinguishing characteristic (višeşa-lakşaņa) of the jīva. The Commentary adds that upayoga is a 'modification', or 'transformation' (pariņāma) of the jīva, which has the form of a function or mode of being of the substance (dravya-vrtti-rūpa).
This is consistent with texts such as Tattvārtha Sūtra
68 Upadhye, Index, p. 36.
69 Upadhye takes it as a dvandva compound: the jīva is 'constituted of sentiency and manifestation of consciousness'.

Page #128

114 Harmless Souls 2.8 (see above), as is the remark that the guņa (the quality - i.e. unique quality) of the ātman is upayoga (gunovaoga tti appaņo).70 Pravacanasāra 2.63 reads:
The self's self is manifestation of consciousness; manifestation of consciousness is said to be knowledge and perception; the manifestation of consciousness of the self is either auspicious or inauspicious.
Again we see Kundakunda repeating the orthodox doctrine of, for example, Tattvārtha Sūtra 2.9, that upayoga consists of jñāna and darśana, and the reformulation of the doctrine that upayoga is the distinguishing characteristic of the self in the epigram that it is the 'self's self (ātmā upayogātmā). A new element is then introduced when upayoga is defined as being either śubha or aśubha ('auspicious' or 'inauspicious'). However, no mention is made of śuddhopayoga here. In the light of the full upayoga doctrine, where śubha and aśubha are both classified as aśuddha, this requires some comment.
Amrtacandra in his commentary on the gāthā (2:63) ignores this omission of suddhopayoga and takes the standard line that upayoga is divided into two, pure and impure, 71 and that of these two, while the pure is free from attachment (niruparāga), the impure has it (soparāga). The latter is of two kinds, śubha and aśubha, corresponding to the twofold nature of attachment, which takes the form of virtue (viśuddhi) or affliction (samkleśa). In the context of the Pravacanasāra as a whole, and of the next gāthā in particular, this expansion is quite natural. However, taken in isolation, Pravacanasāra 2.63 presents a clue as to the origins of 'Kundakunda's' upayoga doctrine. I would suggest that this gāthā reflects, or is part of, an earlier
70 Pravac. 2:42. 71 upayogo dvedhā višisyate śuddhāśuddhatvena - TD on 2:63..

Page #129

Kundakunda: The Pravacanasāra 115 strand of doctrine, one which arose out of or was developed to accommodate lay aspirations, and that the division of upayoga into śubha and aśubha was one of the tools of this development. In brief, beneath the karmic net which characterises life-in-the-world, such a doctrine makes for a limited or impermanent hierarchy of goals and rewards, measured by accumulated amounts of merit and demerit (punya and pāpa) which are associated with particular types of consciousness (śubha and aśubha) (an association which will be examined in more detail when I come to consider the precise mechanism of bondage). For the śramaņa, however, whose ideal goal is total liberation from the karmic net, and thus from rebirth, consciousness directed into the world is at best a potentially dangerous irrelevance. Subha and aśubha punya and pāpa, are equally binding for the ascetic, whose aim is to stop all 'action' (the instrument of karman), even the action of consciousness insofar as it is manifested or directed. Such a reading is borne out by the fact that śuddhopayoga, when it comes to be posited in the full doctrine as the consciousness of pure ascetics, is (as one would always expect from a pollution-free state) viewed in negative terms, i.e. the term designates the absence of aśuddhopayoga. As Upadhye puts it, śuddhopayoga 'is not a positive spiritual something, but only immunity from the remaining two upayogas', śubha and aśubha, and 'upayogas are manifested by the soul because of its being associated with karmic matter'.72 In other words, śuddhopayoga characterises the soul as it is in itself, free of karmic accretion. This seems to be the state which is intimated but not named or given any technical designation at Pravacanasāra 2.64:
72 Upadhye p. Ixxiii.

Page #130

116 Harmless Souls
If the application of consciousness is auspicious the soul accumulates merit; or if inauspicious, demerit; when there is neither (śubha nor aśubha ) there is no accumulation.
Despite Amộtacandra's commentary and Upadhye's remarks to the contrary,73 upayoga as a technical term seems to apply here only to śubha and aśubha. Nothing is said about a condition totally free of karman; it is merely stated that in the absence of subha- and aśubha-upayoga there is no (further) accumulation. Nevertheless, it is quite natural to infer from this a kind of neutral or non-applied state of consciousness. And one may guess that when Jaina scholastics came to consider this doctrine of śubha- and aśubha-upayoga logically, they were led to posit a further category - śuddhopayoga - which was characteristic of developed śramaņas. Technically, it would seem that such a pure state of consciousness is not upayoga at all; i.e. it is not a manifestation or application of consciousness but pure consciousness itself, the parināmika bhāva (of Tattvārtha Sūtra 2.1) in which the jīva 'experiences' kevalajñāna. Thus, theoretically, it should make no difference to the behaviour of ascetics, since there is nothing additional which they must do in order to achieve it; i.e. they must continue as always to avoid the accumulation of new karman and to burn off the residue of 'old' karman. However, the fact that the instruments of bondage for lay people (śubha- and aśubha-upayoga) are seen as internal - states of consciousness rather than physical actions - means that any logical development of the doctrine will be in terms of 'internal' states rather than external behaviour. And this tendency was no doubt aided by the fact that, by the time it was developed logically, bondage for ascetics was already conceived of as predominantly the result of internal attitudes or 'passions' (as in the kaşāya doctrine,
73 Upadhye p. Ixxii.

Page #131

Kundakunda: The Pravacanasāra 117 classically formulated in the Tattvārtha Sūtra).
That the upayoga doctrine represents the culmination and logical end of a lengthy process of 'internalisation' regarding the mechanisms of bondage and liberation, I shall argue at greater length below. Here, I wish to point out that although, technically, śuddhopayoga is the absence of something and not a thing or condition to be attained as such, and that theoretically its articulation should make no difference to the behaviour of śramaņas, in practice it quickly begins to be conceived of in a positive way. And once it becomes something to be achieved, then ways other than simple avoidance of aśuddhopayoga can be prescribed for achieving it. Moreover, its internal' or psychological nature, is bound to be reflected in the methods enjoined for its attainment. Thus in the Pravacanasāra the development of knowledge (jñāna) of the true nature of the self its essential purity), and its realisation through meditation, come to be seen as the quintessential soteriological activities of the śramaņa. The radical implications of this, both for the institution of the renouncer and for ascetic behaviour as previously prescribed, and the reasons why such implications were tempered or largely avoided in practice, will be discussed below as the appropriate passages from the Pravacanasāra are examined in detail.
A full statement of the upayoga doctrine first appears in the Pravacanasāra at 1.9:
When the jīva, whose nature is (subject to) modification, modifies (itself) through auspicious or inauspicious (states of consciousness), it becomes auspicious or inauspicious, then through the pure it becomes pure.
According to this gāthā, the jīva modifies itself (i.e. its soteriological state) through the development of three different states of consciousness, viz. auspicious (śubha),

Page #132

118 Harmless Souls inauspicious (aśubha), and pure (śuddha), and it becomes these states to the degree that its consciousness is them. That is to say, according to this view the jīva is not essentially different from its modification (pariņāma). So, as the Tattvadīpikā (on Pravacanasāra 1.9) remarks, when the ātman (= jīva) modifies itself through the bhāva of śubha or aśubha rāga (passion or attachment), having modification (pariņāma) for its svabhāva, it becomes śubha or aśubha, like a crystal modified by the colour of a red (rāga) contiguous object. Similarly, when it modifies itself through a pure, passionless state, 74 it becomes suddha, like a crystal modified by a pure colour (in effect, an unmodified crystal).
What then are the (soteriological) conditions or goals connected with these states or modifications of consciousness? At Pravacanasāra 1.11 we read:
When the self which has modified itself through dharma is joined to pure consciousness (sampayoga), it attains the bliss of liberation (nivvāņa), or if joined to auspicious (consciousness) it attains heavenly bliss.
Two stages seem to be implied here. First, it is necessary to modify oneself through dharma. Dharma has been defined (at Pravacanasāra 1.7) as 'conduct' (căritra), which could hardly be argued with from the orthodox Jaina point of view; however, in the same gāthā, dharma is further defined as sama (sama), equanimity, and sama as the pariņāma of the ātman in which it is free from the disturbance of moha (delusion).75 Dharma is thus not so much linked with cāritra in the sense of physical activity in the world, as with a particular state of mind, sama (Sk. śama, 'tranquillity', 'absence of passion'), connected here
74 suddhenārāgabhāvena - TD on Pravac. 1:9. 75 mohakkhohavihīņo - Pravac. 1.7 (= Sk. mohakṣobhavihinah).

.

Page #133

Kundakunda: The Pravacanasāra
119
specifically with the absence of moha, and probably attained through some form of preliminary meditation or 'calming'.76 The second stage involves the development of śuddha or subha states of consciousness, one resulting in final liberation (nirvāṇa), the other in rebirth in a heavenly world. This at least is the sense if we take sampayoga as a synonym for uvaoga, as the context would suggest. This is what the commentary (Tattvadipikā) does, in the process warning against the (heavenly) bondage attendant upon subhopayoga and enjoining śuddhopayoga instead.
The following gathā (1.12) lists the inevitable (de)gradations of the atman through the rise of the inauspicious manifestation of consciousness. This is contrasted (at 1.13) with the 'happiness' (suham / sukham) of those who have attained suddhopayoga, providing the first unambiguous reference in the text (as opposed to the commentary) to the technical term suddhopayoga.
In the light of what follows, the description (at 1.14) of the śramana whose upayoga is suddha as one who has understood well the padarthas (i.e. the nine tattvas) and the sūtras, who is endowed with self-control (samyama) and tapas, who is free from desire (vigatarāga), and to whom sukha and duḥkha are alike, seems more like the attempted elision of a received 'external' tradition with a more recently formulated 'internal' one than a description of the ideal embodiment of suddhopayoga as such.
Gāthā 1.15 emphasises the autonomy of the atman
whose upayoga is pure: it becomes itself the self77 because it is free from all forms of destructive karman.78 Having reached this state, it is omniscient and known as selfexistent (sayambhu / svayambhữ) (1.16). The point to note
76
For comments on Jaina meditation, see p. 185ff., below. 77 bhūdo sayam evādā - Pravac. 1.15 (= bhūtaḥ svayam evātmā). 78 See JPP p.
115ff.

Page #134

120 Harmless Souls here is that this journey to omniscience, to total purity, is seen as the recovery or realisation of the 'real' or 'true' self, i.e. of its essentially unbound nature.79
Svayambhū (self-existent) is a resonant term in Indian religious thought, and it is probably not coincidental that the text here uses an expression with a decidedly monistic and Vedāntic ring to it, one which is used in various contexts as an epithet for the highest principle.80
Returning to the condition of those who are still characterised by aśuddhopayoga, we read at 1.46 that:
There would be no samsāra for any embodied jiva if the atman by itself, through its own nature, did not become auspicious or inauspicious. 81
In other words, bondage - life in samsāra- is the direct result of particular states of consciousness which are selfgenerated. Thus the ātman does not find itself bound, its condition in the world is not existentially 'given'; it has bound itself. As the commentary (Tattvadīpikā) puts it:
79 That it has this meaning, despite the standard Jaina doctrine that the self has been bound in matter from beginningless time (see JPP p. 107), will be made clear below. As the commentary (TD) on this gāthā (1:16) remarks, from the niscaya point of view - i.e. the determined or higher, as opposed to the conventional (vyavahāra) view (see p. 126, fn. 5, below) - there is no causal or instrumental relation between the self and anything else, viz. the ajīva category, including material karman (na niscayataḥ pareņa sahātmanaḥ kärakatvasambandho 'sti). : 80 See, for example, Brhadāranyaka Upanişad 2.6.3.: brahma svayambhū brahmane namaḥ. I shall have much more to say about the condition of the śramana whose upayoga is suddha in the sections on the mechanisms of bondage and liberation.
81 It should not be thought that I am taking these gāthās out of context. Gāthā 1:46 provides a good example of the disjointed nature of parts of the Pravacanasāra, for it is sandwiched between a gāthā on the activities called ksāyiki (due to the destruction of karmas) and one on the knowledge called ksāyika (produced after the destruction of karmas).

Page #135

Kundakunda: The Pravacanasāra 121
If the ātman itself is not transformed exclusively by its own nature, consisting of auspicious and inauspicious states, then it remains with its own pure state, without division, in every way and forever. 82
This is strikingly at odds with the standard Jaina doctrine, summarised by P.S. Jaini as follows: the soul's 'involvement with the material universe ... has had no beginning, and it is likely to continue almost indefinitely'; furthermore, the Jaina 'believes it is incorrect to imagine that the soul was once pure but later became defiled. It has always been impure, just as a seam of gold has "always" been imbedded in the rock where it is found'.83 Gāthā 1.46 implies a different view - that the natural or 'original' condition of the soul is unbound. In other words, Kundakunda denies that it is an ontological given that the soul is entrapped in matter from beginningless time, without initial cause; instead, he states that the cause of bondage (being in the state of samsāra) is the soul's developing by itself into śubha and aśubha states. And once an original cause for bondage is posited, it is possible to go beyond that and describe the state before bondage, and this becomes the original or natural condition of the soul, i.e. the state from which each individual jīva has started and to which it aspires to 'return' - in this case śuddhopayoga. Furthermore, if from the individual's point
82 yadi khalv ekāntena śubhāśubhabhāvasvabhävena svayam ātmā na parinamate tada sarvadaiva sarvathā nirvighătena śuddhasvabhāvenaivāvatişthate - TD. on Pravac. 1.46.
As we shall see, in the Samayasāra Kundakunda actually affirms this view: from the niścaya standpoint, the soul is, in every way and forever, in its own pure state. That naya is clearly not the one taken here - i.e. the soul does in some real sense transform itself - however, some important doctrinal modifications (noted below), make this gāthā one signpost on the road to such a position.
83 JPP p. 107.

Page #136

122 Harmless Souls of view samsāra is self-generated by the ātman through its manifestations of consciousness, this not only implies that the ātman is prior to samsāra but also that it 'creates' samsāra through its own states of consciousness (i.e. 'creates' it in the only significant sense - for itself). Taken in this way, samsāra would thus be the creation of impure or 'false' states of consciousness. Given the essential purity of the soul, however, this false consciousness can only take one form, namely, delusion regarding the ātman's true nature. This is indeed the case in the Pravacanasāra, where moha comes to be seen as the chief agent in bondage, and is directly linked to aśuddhopayoga (śubhaand aśubha-upayoga).84 This, as will be made clear, has radical implications for the prescribed methods of liberation, entailing release through inner transformation of consciousness and knowledge (gnosis). But before considering these, I shall now discuss the mechanism of bondage as portrayed in the Pravacanasāra in greater detail.85
84 For a full discussion of this, see p. 124ff., below.
85 Whether it is possible logically to hold a view that the world - samsāra - is a product of false consciousness, and that false consciousness is produced by various obscuring karmas which are themselves the material constituents of samsāra, is not a problem directly confronted in these texts. However, the implication of Kundakunda's niscaya view is that false consciousness is entirely selfgenerated and thus really has nothing to do with karman at all. The vyavahāra view, of course, is different; and that is enough, according to the orthodox Jaina philosophical position, to neutralise the difficulty. For a historian, however, the fact that contradiction is built into the system explains nothing; rather it is necessary to explain why such a system, with its deliberate imperviousness to argument, should have been erected in the first place. Kundakunda's 'logical' (niscaya) view, by demonstrating the position Jaina scholastics would have found themselves in without the safety-curtain of the various nayas, helps us to formulate such an explanation. For, as we shall see when we examine the Samayasāra, he remained largely true to the logic of his own argument by employing the niscaya / vyavahāra distinction in a

Page #137

Kundakunda: The Pravacanasāra 123
non-'contradictory', and thus, according to his critics within Jainism, one-sided and heretical way.

Page #138

The mechanism of bondage according to the
Pravacanasāra
5.1 The mechanism of bondage
The principal passages dealing with the mechanism of bondage in the Pravacanasāra are clustered in a group in Book 2 (c. gāthās 2.77-2.97). The pivotal gāthā here is Pravacanasāra 2.81, where the question is asked, how is it possible for the non-material ātman to be bound by material karman. The text in full reads:
The embodied (mūrta, i.e. material objects), having qualities such as rūpa, etc., is bound by mutual touching; the ātman is quite different from [the opposite of] that, so how can material karman be bound to it?
In many ways this is a crucial question for Jaina dogmatics, and no less so for Kundakunda. It was one which was probably quickly identified by opponents in rival darśanas as a weak point in the Jaina argument; and as we shall see, it may well have been Kundakunda's attempt to solve this specific problem, although perhaps more for 'internal' than 'external reasons, which first put him on the path that ends in the apparently heterodox views of the Samayasāra.
To understand the importance of the problem we must look at the way the argument is developed in the Pravacanasāra in some detail. At 2.77 it is said that molecules (khamdha / skandha), 'capable of becoming karman', coming into 'contact' with, or 'attaining to' a transformation of the jīva, develop into the state of karman (kammabhāva); but it is not by the jīva that those skandha are transformed. In other words, it is neither the jīva nor the skandha which is directly instrumental in the 'creation'

Page #139

Kundakunda: The Pravacanasāra 125 of karman (and thus bondage), but the meeting of the two, i.e. of the jīva's condition, or mode at the time, with the molecules (pudgala-skandhas). However, as the commentary (Tattvadīpikā) makes clear, it is the skandhas which have the potential to transform into the karman state (karmabhāva), even without the jīva side of the equation. So the soul (puruşa) is not the agent of the karman-state of material masses (pudgalapiņda). Thus the creation of karmic bondage is defined here as the transformation of material molecules by 'contact with the modifications of a jīva, with 'contact' and transformation, and not the jīva, being the instrumental factors as such. Clearly, the particular modification of the jiva also plays a crucial role, one we shall consider later, although it is not emphasised here.2 It is this material, transformed into karman, which goes to make up the various bodies the jīva is bound into in its various births. However, as the commentary reminds us, the jiva, since it is not material, cannot be - and therefore should not be identified with - any of these bodies.4
The soul's characteristics are then enumerated:
Know that the jiva, whose quality is consciousness (cetanā), is without taste, without form ('colour'), without odour; it is unmanifest (i.e. cannot be touched), without sound, beyond the range of a distinguishing or characteristic mark, and without describable configuration. [Pravacanasāra 2:80]
This is clearly a description of the jīva in its unbound or pure state (i.e. in Kundakunda's terms it is the niscaya view
I na pudgalapiņdānām karmatvakartā puruṣo 'sti - TD on Pravac. 2:77.
2 What is meant by 'contact' in these circumstances, and thus the reason for its being placed between inverted commas, will be clarified below.
3 Pravac. 2:78. 4 TD on 2:79.

Page #140

126 Harmless Souls of the jīva).5 Nevertheless, there is something odd about such a gāthā in a Jaina context. Why, for instance, is the jīva said to have no definable structure or configuration, when standard Jaina doctrine is that it has the shape of its current or, if released, final body?6
I shall have more to say about this later, but it is unlikely that it is simply a coincidence that the characterisation of the pure jīva given above could be just as well a description of the Vedāntic ātman-brahman, even down to the fact that it has cetanā as its quality (guna). In this respect it is interesting to compare Kundakunda's characterisation with that given in, for instance, Nemicandra's Dravya-samgraha, a tenth century Digambara work referring back to the Șatkhandāgama and
5 Matilal points out that Kundakunda's dual classification (vyavahāra- and niscaya-naya) 'has no direct connection with the usual seven standpoints of the Jainas, but corresponds to the ... distinction of two levels of truth in Mādhyamika Buddhism' or the same distinction made in Sankara's school of Advaita Vedānta. This view, which is peculiar to Kundakunda and those following him, describes the soul from the niścaya standpoint 'as independent, self-existent and uncontaminated by matter. This is the truth in the ultimate sense, a goal to be arrived at the final stage (sic.)'. But the vyavahāra standpoint 'describes the soul as one that is involved in karma as well as in the birth and re-birth cycle (samsāra)' (1981, p. 43).
I shall discuss this distinction and the ways in which Kundakunda uses it at length when I deal with the Samayasāra - see p. 239ff., below.
Samsthāna is the term used, for example, at TS. 8.11 to denote the figure of the body. There are, however, other instances of this kind of description; thus in a passage in the Ayāramga-sutta the liberated self is held to be indescribable: arūvi sattā, apayassa payam natthi - 'its essence is without form; there is no mark for what is without mark' (quoted by P.S. Jaini, JPP p. 271, fn. 41., who translates, 'there is no condition of the unconditioned').
On the dimensions of the jiva, see JPP pp. 58ff., 102, 269. As P.S. Jaini remarks, it is only the Jainas who posit a soul that is at the same time 'nonmaterial and yet subject to contraction and expansion when in its mundane state and is therefore 'of the same dimension as its body (sva-deha-parimāņa)' [JPP p. 58). So this is part of the same problem as that raised at Pravac. 2:81.

Page #141

Kundakunda: The Pravacanasāra 127 the Kasāyaprābhrta, the only works designated canonical by the Digambaras. There, a kind of composite view of the soul is given which includes the characteristic of its being the size of its body; and when the niscaya view is raised it simply designates the soul as that which has cetana:
It (the soul) is (1) jiva (that which lives); (2) possessed of upayoga ...; (3) amūrta (immaterial); (4) kartā (the doer of all actions); (5) sva-deha-parimāņa (of the size of its body, which it completely fills); (6) bhoktā (enjoyer of the fruits of actions); (7) samsāra-stha (located in the cycle of death and rebirth); (8) siddha (in its perfect condition a siddha); (9) ürdhvagati (of an upward tendency). That which in the three times has four prāņas, viz. senses, power, vitality, and respiration, is conventionally soul; but from the essential point of view that which has consciousness is soul.?
The term 'niscaya' is used here merely to designate the focused or narrow view of the essential characteristic of the soul; and unlike the niscaya view given at Pravacanasāra 2.80, it does not contradict the vyavahāra, or conventional Jaina view. That the niścaya-naya of the Pravacanasāra is apparently irreconcilable to the conventional naya does not, of course, disqualify either according to Jaina 'logic'; however, I shall have more to say about this when I discuss the possible reasons for the development of the naya doctrine and Kundakunda's unconventional, not to say 'heretical' reading of it, below.
7 jīvo uvaogamao amutti kattā sadehaparimāņo
bhottā samsārattho siddho so vissasoddhagai || 2 || tikkāle cadu pāņā imdiyabalamāu āņapāņo ya|
vavahārā so jivo nicchayanayado du cedanā jassa || 3 || Nemicandra's Dravya-samgraha, quoted by J.L. Jaini 1940, p.83; his translation with alterations.
I am unable to translate vissasa, but it is probably connected with (Sanskrit) visvañc, 'going in all directions'. That is to say, the soul, depending upon its karmic condition, is capable of going in any direction on the death of the body, although its inherent tendency is upward. It may be that vissasa is a corrupt reading.

Page #142

128 Harmless Souls
Returning to the Pravacanasāra, we come again to the crucial question (at 2.81), how can the immaterial soul be bound by material karman? The fact that this question is being asked at all is highly revealing of the shift in doctrinal emphasis, one which, as we shall see, entails or accompanies a shift in practice. For according to standard Jaina doctrine there is no need to make such a query, since it is an existential fact - a 'given' - that the jiva has been entrapped in matter from beginningless time. As we have noted above, and as P.S. Jaini points out, the Jaina envisions his soul's 'involvement with the material universe' as having had no beginning. Furthermore, he 'believes it incorrect to imagine that the soul was once pure but later became defiled. It has always been impure ... !8 Thus, the received doctrine points to a question which is more practical than theoretical, viz. how can what is impure be made pure?
If, however, you try to account for the fact that the jiva is, or appears to be bound, first you must assume some primal or 'original' state in which it is not bound, and then attempt to show what is demanded by this particular question (Pravacanasāra 2.81), namely, how is it that the pure can become, or appear to be, impure? This is easy enough to answer when the two have the same ontological basis (a white sheet can be covered in mud), but when they have no such common ground, one being material and the other not, then a logically satisfactory answer such as might silence an opponent is going to be much harder to formulate. As the Tattvadīpikā (on 2.81) puts it: with embodied, material substances (mūrtayor ... pudgalayor) which have rūpa, etc., mutual bondage due to particular touch (i.e. to the state of being smooth or rough / sticky or dry (snigdha or rūkşa]) is recognised;9 but how can the ātman, being incorporeal (amūrta), and so without
8 JPP p. 107.
9 snigdha-rūkşatva-sparśaviseșād-anyonyabandho 'vadhāryate - TD on 2:81.

Page #143

Kundakunda: The Pravacanasāra
129
particular contact or touch, stickiness or dryness, be connected with material karman (karmapudgala) which has those qualities?
To see how Kundakunda (and Amṛtacandra) attempt to answer this question, we must look in detail at the gāthās following Pravacanaṣāra 2.81.10
At 2.82, we are told that although the jiva is without rūpa, etc. (as enumerated at 2.81), itself, nevertheless, it sees and knows (pecchadi jāṇādi) substances (dravya) which have rupa, etc., and that the mechanism of bondage is analagous to this. Thus the jiva, which consists of upayoga, having attained to various objects (of the senses), is infatuated, attached or averse, and so it is combined with them.11
The Tattvadipikā (on 2.83) explains that the atman, composed of upayoga, meeting with various distinct forms / objects, betakes itself (samupaiti; sam-upa- vi) to moha, rāga, or dveṣa. And because the innate nature of the atman (ātmasvabhāva) is 'coloured' (uparakta) by moha, etc., which are conditioned or modified by the 'other' (parapratyaya) (i.e. the ajīva, the 'various objects'), like a crystal gem whose nature is coloured by the objects close to it, so the atman becomes itself 'bondage' (bandha) because it is paired with their natures (tadbhāvadvitīyatvāt).
The gist of this seems to be that passions are generated in the atman by proximity to objects, that such passions are modifications of the upayoga which characterises the atman, and that this is what constitutes bondage. In other words, the ātman binds itself through its own upayoga.
On the one hand, this gāthā and its commentary seem to be harking back to earlier ideas about the soul being
10 Since there is a good deal of repetition in these verses, in what follows I shall rely more on summary than direct quotation; the relevant gāthās are Pravac. 2:81-2:97.
11 Pravac. 2:83. There is some ambiguity here over whether tehim refers to the sense-objects or to the passional states; Upadhye translates it as referring to the latter, although it amounts to the same thing.

Page #144

130 Harmless Souls
'coloured' (the leśya doctrine) which strongly suggest that the soul was once viewed in material terms, while on the other hand, the very idea of bondage seems to have been lifted out of the sphere of material karman altogether: the soul now binds itself through modification of its manifestation of consciousness. 12 Such modifications are said to be brought about by contact with or proximity to material objects; yet the mechanism of this is not explained; at the level of contact the problematical gap between the material and the immaterial remains. However, it looks very much as though bondage has in effect been dematerialised. For if it is the particular manifestation of the immaterial soul's own consciousness which is instrumental in bondage, what place is there for the original karman theory?13 The only answer is that it is the particular upayoga which 'causes' the karmic particles to adhere to the jīva (i.e. upayoga has the same role as kaṣāya does in the classical theory), but this only returns us to the original problem: how can what is immaterial - a state of consciousness characterising an immaterial soul - be instrumental in its own bondage by matter? A close reading of Kundakunda and Amṛtacandra suggests that they were aware of these problems. In particular, the epistemological divide between the vyavahāra and the niscaya views, with Kundakunda's peculiar interpretation of this division (which becomes so evident in the Samayasāra), provide them with a possible route out of this impasse. But as we shall see later, this way offers such a threat to the whole tradition of Jaina doctrine and practice that, with the exception of the Samayasāra, it cannot be
12
On lesya see, for instance, Utt. XXXIV and Jacobi's footnote on Utt. XXXIV.1 (1895, p. 196) which quotes the same metaphor of the crystal (sphatika) from the Avacuri with regard to the lesya's influence on the soul. See also Schubring (1962) para. 97-98.
13 Although, as we shall see below, Kundakunda and Amṛtacandra are reluctant to admit this: according to the niścaya view, jīva is the instrumental cause of material karman, and the material cause of its own modifications.

Page #145

Kundakunda: The Pravacanasāra
openly entered into or allowed to go beyond implication. Pravacanasara 2.84 confirms that it is by the (internal) state or attitude (bhāva) with which the jiva sees and knows objects that it is stained (rajjadi), and it is this state which is instrumental in bondage. According to the commentary (Tattvadīpikā), this bhāvabandha corresponds to, or represents, snigdha and rūkṣa, i.e. it takes the place of the material instruments of bondage. Thus dravyabandha is controlled or modified by bhavabandha.14 To put it simply, bhāva is what counts in bondage, dravya cannot be bound without it. 15
131
At Pravacanasāra 2.85 it is reiterated that whereas bondage between material atoms takes place through touching (sparśa), bondage of the jiva takes place through passion (rāga, literally, the act of colouring); moreover, pudgala and jiva are said to be bound together by 'mutual interpenetration' (annonnassavagāho). As the next gāthā (2.86) makes clear, this last part refers to the standard Jaina doctrine that pudgala and jīva / ātman do not touch in bondage but occupy the same space:
The self has space-points; material bodies enter into those spacepoints. They remain there as fit; they go [or] they are bound.
This takes us to the heart of the problem, succinctly diagnosed by P.S. Jaini when he explains that:
Jainas view the soul's involvement with karma as merely an "association" (ekakṣeträvagaha, literally, occupying the same locus); there is said to be no actual contact between them, since this would imply a soul which was, like karma, material by nature. Just how a non-material thing can in any way interact with a material one is not well clarified. The texts simply suggest that we can infer such an association from our own "experience" of
14
atha punas tenaiva paudgalikam karma badhyata eva, ity eșa bhāvabandhapratyayo dravyabandhaḥ - TD on Pravac. 2:84. 15 Cf. TD introducing 2:87, below.

Page #146

132 Harmless Souls
bondage, just as we infer the association of an immaterial consciousness and a material object from the experience of perception.
He then refers to the passage from the Pravacanasāra and the Tattvadīpikā (2.82) which we have noticed above. 16
The attempts of the Tattvadipikā to explain the fact of bondage, given these conditions, merely circle the conundrum without engaging it (and it is difficult to imagine what solution could be offered to a problem expressed in such terms). Thus the Tattvadipikā on Pravacanasāra 2.86 'explains' that the ātman has innumerable space-points (pradeśa), equal to those of the inhabited universe (lokākāśa);17 within those pradeśas the soul vibrates, depending on the 'atom groups' (vargaņā) of body, speech and mind. Karmic material, having a similar vibration (parispanda), enters the pradeśas and remains there. Then, if the soul is in a state (bhāva) of moha, rāga and dveșa, it is also bound. (In addition to the problem of how a bhāva brings about material bondage, here there is also the question of how material atoms can 'cause the immaterial soul to vibrate.) The Tattvadipikā on Pravacanasāra 2.85 claims that the modifications (paryāyas) of rāga, moha and dveşa are limiting adjuncts (upādhikas) in relation to the jīva, and that their oneness with the jiva constitutes bondage. Moreover, the bondage (interpenetration of jīva and karmapudgala is merely the occasion (nimitta) of their mutual modifications, i.e. they modify themselves, not each other. 18 Quite how this process is set off is again not explained.
The commentary (Tattvadīpikā) on Pravacanasāra 2.82 related this directly to the upayoga doctrine, explaining
16 JPP pp. 113-114. 17 On this, see JPP p. 102.
18 jīvakarmapudgalayoh parasparapariņāmanimittamātratvena visistatarah parasparam-avagāhah sa tadubhayabandhah - TD on Pravac. 2:85.

Page #147

Kundakunda: The Pravacanasāra
133
that, although the self, because it lacks touch due to its absence of rūpa, has no union with karmapudgala, the vyavahāra view of bondage by karmapudgala is 'proved' (sādhaka) by the union with states of rāga, dveșa, etc., 'based upon a manifestation of consciousness conditioned by karmapudgala existing in the state of being immmersed in the same space [as that occupied by the self]'.19 In other words, karmapudgala, occupying the same pradeśas as the jīva, brings about (subha or aśubha) upayoga (rāga, dveṣa and moha), which is instrumental in bondage. Thus, according to the vyavahāra view, it is still material karman, or dravyabandha, which 'causes' bhāvabandha (particular manifestations of consciousness); so it is karmapudgala which is ultimately responsible for bondage. The niscaya view here, on the other hand, is that it is bhāvabandha (upayoga, etc.) which is the cause of dravyabandha and thus the initiating cause of bondage. That is to say, it is the self-transformation of the jiva into states of upayoga characterised by rāga, etc., which 'sets off the transformation of karmapudgala and brings about bondage. (I shall have more to say about this below.) This niscaya view is given in the Tattvadīpikā at 2.86, and again in the introduction by Amṛtacandra to the next gāthā (2.87).20 The actual text (Pravacanasāra 2.87) is less technical but quite clear:
The impassioned self binds karman, the self free from passion is freed from karman: this is the summary statement of the bondage of souls know this in reality.
The Tattvadīpikā comments that the jīva free from rāga
19 ātmano nirūpatvena sparśaśūnyatvān na karmapudgalaiḥ sahāsti sambandaḥ, ekāvagāhabhāvāvasthitakarmapudgala
karmapudgala
nimittopayogādhirūḍharāgadveṣādibhāvasambandhaḥ bandhavyavahārasādhakas tv asty eva - TD on Pravac. 2:82.
20 dravyabandhasya bhāvabandho hetuḥ - TD on 2.86. atha dravyabandhahetutvena rāgapariṇāmamātrasya bhāvabandhasya niscayabandhatvam sādhayati - TD intro. to 2.87.

Page #148

134
Harmless Souls
not only fails to bind fresh (abhinava) karman, but it also rids itself of old karman, and so liberates itself (i.e. absence of rāga functions as both samvara and nirjarā). It is clearly, therefore, modification into attachment (rāgapari ṇāma) which is the 'real' bondage (that is to say, according to this view, it is the prime or ultimate cause of bondage).
What precisely is being claimed here, and what are we supposed to understand by the use of vyavahāra and niscaya in this context? In an attempt to clarify this, I shall construct a theoretical model to test against the text. And it is worth remembering that, freed from the commentary, this section of the Pravacanasara has all the appearance in its present form of being a compilation of gāthās on bondage, and so may represent different layers of doctrine.
First, it should be noted that Pravacanasāra 2.97, which draws the line under this section on bondage, says that this description of the bondage of the jiva has been preached by Arahants for yatis (ascetics) from the niscaya point of view, whereas another point of view, the vyavahāra one, is preached for others (i.e. the laity). In other words, this is a teaching aimed specifically at śramaņas.
The vyavahāra view, which is not preached here, is, as we have seen, outlined in the commentary (Tattvadīpikā) on 2.82. It amounts to this: dravyabandha (karmapudgala) is the cause of bondage, for it is karmapudgala, through occupying the same pradeśas as the jiva, which brings about the modification of consciousness (upayoga) into raga, dveṣa and moha (subha and aśubha),21 which in turn is the cause of further bondage (i.e. it 'sets off the binding of further karman).
The niscaya view is that it is bhāvabandha (subha- and asubha-upayoga) which is the cause of bondage, since, although karmapudgala occupies the same pradeśas as the jīva, there is no contact (sparśa) between the two. Rather, their contiguity 'sets off' in the jiva self-transformations into
21 See Pravac 2·88.

Page #149

Kundakunda: The Pravacanasāra 135 states of upayoga, characterised by rāga, dvesa and moha, and these alone cause karman to bind the jīva.22 (What is meant by 'karman' in this case will be discussed below.)
The content of these two views can be combined or reconciled in a circular model, in function not unlike the pațicca-samuppāda of the Buddhists. That is to say, karman causes transformation of consciousness, causing (new) karman to be bound, causing (further) transformation of consciousness, causing (new) karman to be bound, etc. The vyavahāra and the niscaya views both turn this circle into a chain. The difference between them is that, from the vyavahāra perspective, karmapudgala is the initiate cause of bondage (although strictly speaking dravyakarman is beginningless, corresponding to the standard Jaina view of the predicament of the jīva in samsāra, that it has always been bound), while, from the niscaya perspective, it is upayoga which is the initiate cause.23
The vyavahāra view is spelt out at Pravacanasāra 2.29:
The soul, tainted by karman, attains to a modification which is connected with karman; thus karman adheres; therefore karman is a modification (of the soul).24
The Tattvadīpikā on this gāthā explains that the modification (pariņāma) of the ātman is the cause of material karman adhering (dravyakarmaśleşahetuḥ), and the cause of this pariņāma is itself dravyakarman. This is not, however, the 'fault' (doşa) of 'mutual dependence' (itaretarāśraya) because the ātman, which is bound by beginningless dravyakarman, employs as its cause the prior dravyakarman.25 And because new and old material
22 See Pravac. 2:87.
23 The precise meaning of karman in the niscaya chain will be discussed below.
24 Upadhye adds, 'developed by passions etc.'. 25 anādiprasiddhadravyakarmābhisambaddhasyātmanah
prāktanadravyakarmanas tatra hetutvenopādānāt - TD on

Page #150

136 Harmless Souls karman are effect and cause, then the modification of the self is dravyakarman, and the ātman, because it is the agent of its own modification, is figuratively (also) the agent of dravyakarman.26 Here the emphasis clearly falls on material karman as the 'real' or 'first' cause of bondage, i.e. the cause which one must concentrate on removing to achieve liberation.27
The standard explanation of these two views is given in the Tattvadīpikā on Pravacanasāra 2.97, where it is claimed that both are correct, since the dravya jīva is thought of as both pure and impure.28 For according to the standard nayas doctrine,29 these are two ways of expressing the same thing - from the point of view of substance (dravyārthika) and from the point of view of modification or mode (paryāyārthika). So from the former the jiva is pure, from the latter it is impure. Here, according to the Tattvadīpikā on 2.97, the purity of the jīva is emphasised because it is the most effective' or 'conclusive' (sādhakatama). In other words, as a pedagogic principle, either the purity or impurity of the jīva is stressed, depending upon which audience is being addressed. To use a Buddhist term again, this might be seen as a 'skilful means' to engender the type of behaviour most required at a particular rung on the soteriological ladder. However, from a logical point of view (i.e. the point of view required in debate) this brings us no closer to answering the question raised at Pravacanasāra 2.81, namely, how can the immaterial soul be bound by, or bind, material karman? Nor does it answer the variant on that question, raised here
Pravac. 2:29. 26 tathātma cātmapariņāmakartstvād dravyakarmakartāpy
upacārāt - ibid. 27 Cf. the niscaya view at Pravac. 2:30, where the jiva is said not to be the agent of material karman; rather its action in transforming itself is viewed as karman, therefore it binds itself.
28 ubhāv apy etau staḥ śuddhāsuddhatvenobhayathā dravyasya pratiyamānatvāt - TD on Pravac. 2:97.
29 See, for instance, J.L. Jaini 1940, p. 116.

Page #151

Kundakunda: The Pravacanasāra 137 by the niścaya view: how can what is essentially pure transform itself into something impure? To say that at the same time it is and is not bound by material karman, depending on one's point of view, is to side-step the
question.
To be clear about this, the naya view is not the tautological one that, from the standpoint of the liberated jiva, there is no bondage by material karman, while, from that of the unliberated soul, there is. Rather, it is saying that at one and the same time the jīva is bound and not bound by karmapudgala, depending on one's point of view. This is in many ways a risky concession; to say that there is a viewpoint from which the jīva is unaffected by material karman, here and now in samsāra, even if the purpose of such a statement is primarily prescriptive and is not supposed to be a complete view, is to open a vista upon a potentially very different soteriological route. It is this path, as we shall see, that Kundakunda takes in the Samayasāra through his use of a vyavahāra / niścaya distinction which has more in common with Madhyamika Buddhism and even more with Advaita Vedānta than with the Jaina philosophy of Anekāntavāda. Here, we can come closer to understanding the origins of that departure by examining further the implicit tendency of this section of the Pravacanasāra on the mechanism of bondage.
As we have seen, the niscaya-naya is one attempt to break out of the logical impasse created by the material bondage of a non-material soul. It suggests that, at some level, the soul binds itself, which, if it is non-material, would seem to be the only intelligible explanation for the fact of bondage. (Although this, of course, raises other theoretical problems.) However, Jainism as a religion, as opposed to a philosophy, needs at the same time to keep the connection with material karman, which is the basis of all its ethical and ascetic practices. This connection is provided by the vyavahāra-naya. As an ungenerous opponent might remark, the naya doctrine seems like a Jaina attempt both to have their material karman and to

Page #152

138 Harmless Souls destroy it. It might be objected here that karman, although not an initiating cause, is still an essential component in the mechanism or chain of bondage according to the niścaya view. To see whether this is really the case, we must make a closer examination of the relevant gāthās in the Pravacanasāra.
As we know, the niscaya view is that when the soul is attached or impassioned (ratta) it binds karman, and that when it is free from rāga it is free from karmas (Pravacanasāra 2.87). Expanding on this, the next gāthā (2.88) states that bondage comes from modification, which consists of attachment (rāga), aversion (dosa), and infatuation or delusion (moha).30 Delusion and aversion are inauspicious, attachment is either auspicious or inauspicious (depending upon whether it takes the form of purification (viśuddhi) or defilement (samkleśa) [Tattvadīpikā]). Pravacanạsāra 2.89 continues:
It is said among ignorant people (annesu) that the śubha modification is merit (punya), the aśubha is demerit (pāpa). The modification that does not result in anything else is, according to the Jain religion, the cause of the destruction of duḥkha.
In other words, while aśuddhopayoga is the cause of punya and pāpa, the cultivation of suddhopayoga leads to liberation.31
Next (at 2.90), it is affirmed that the jīva is essentially different from its embodiments in the six classes of living beings, whether immobile (sthāvara) or mobile (trasa). For as the Tattvadipikā puts it, among these six classes of embodied jīvas, there are other substances (viz. pudgala, etc.), whereas the ātman is its own substance.32
30 Cf. Pravac. 1.84.
31 Compare this with 2:64 (quoted above), which is probably the gāthā which 2:89 refers back to, where it is clearly stated that it is upayoga which accumulates karmic residue (merit and demerit).
32 atra sadjivanikāyātmanah paradravyam eka evātmā sva

Page #153

Kundakunda: The Pravacanasāra 139 We now come to a crucial gāthā [2.91]:
Who does not know thus the paramātman, encountered in their own natures, conceives through delusion the idea 'I am (this), this is mine'.
This gāthā has a number of ambiguities; for instance, paramappāņam apparently refers to the Paramātman - a neo-Vedāntic term possibly used by Kundakunda at 2.102 (see below), and frequently employed in the Tattvadīpikā, especially on Book 1 - rather than to the ajīva and the jiva (which is the way Upadhye takes it). However, the general meaning is clear: whoever does not know the essential nature of the ātman - its difference from its embodiments - comes to have the delusive idea that what is in reality not himself - the ajīva or para - is himself; he confuses two totally separate categories. This cognitive or epistemological confusion about the true nature of the self is, as we have seen (2.88), a self-transformation of the jīva (aśubhopayoga), said to have been 'set off by contiguity with material karman, although the mechanism of the latter cannot be explained (and, as I shall suggest, is not really needed according to the logic of the internalised system).
In short, delusion (moha) is at the beginning of the causal chain of bondage; for if the soul is not really connected with ajīva (including karman) then only moha can 'persuade it to think that it is. However, moha itself, in the standard explanation, is a result of material karman. If this is how Kundakunda views moha - i.e. if material karman is still playing an instrumental or catalytic role in bondage because of the inexplicable reaction of the jīva to its contiguity - then one would expect freedom from moha, etc., and thus from bondage, to result from the destruction or obstruction of material karman. However, according to the Pravacanasāra, it is meditation on the fact of the
dravyam - TD on 2:90. Note the typically Vedāntic distinction here between jīva, embodied soul, and ātman, pure or essential soul.

Page #154

140 Harmless Souls
complete separateness of the pure self, and thus the realisation of that fact, which brings about freedom from moha, and liberation; nothing is said about material karman, all is internal. To illustrate this point, it is necessary to trespass a little on themes I shall consider in their full context and at greater length in a later section, 'The mechanism of liberation'.
There is a clear connection between the description of the jiva subject to moha at Pravacanasāra 2.91 and gāthās 2.98 and 2.99. Pravacanasāra 2.98 reads:
He who does not abandon the idea of 'mine' with regard to body and possessions - (thinking) 'I am (this), this is mine' - gives up the state of being a śramana and becomes one who has resorted to the wrong road.
And 2.99 provides the antithesis to this, i.e. it describes the attitude of the śramana:
He who meditates in concentration, thinking 'I am not others' and they are not mine; I am one (with) knowledge', comes to be a meditator on the (pure) self.
At Pravacanasāra 2.100, the self, which in meditation 'I' know to be myself, is described as being constituted of jñāna and darśana, an object beyond the senses, eternal (dhuva), unmoving (acala), without support (i.e. independent) (aṇālamba), and pure (suddha). Bodies, possessions, happiness, suffering, enemies and friends are not the 'eternal associates of the soul' (jīvassa ṇa samti dhuva);3 ;33 the self whose self is upayoga is eternal (dhuvovaogappago appa), i.e. the pure self (2.101). Knowing this (its true identity), the pure self meditates on the highest self (the paramātman) / the self and the other (the non-self),3 ,34 with formed or formless thought, and so
33 Upadhye's trans., p. 23.
34 TD translates as paramātman, the Prakrit text prints param

Page #155

Kundakunda: The Pravacanasāra 141 destroys the evil (literally, 'very difficult') knot of moha (khavedi so mohaduggamthim) [2.102]. And he who has overcome this knot of delusion, having destroyed rāga and dveşa (which, as the Tattvadīpikā points out, have moha as their root), being indifferent to pleasure and suffering, attains to undecaying happiness (sokkham akkhayam), i.e. liberation (2.103).
This whole passage is, of course, redolent of Vedānta. Here, I simply want to point out that liberation is seen to be attained not by the destruction of that karman which (very tenuously) has been said to bring about moha, but by the destruction of moha itself through meditation on the essential purity and complete separateness of the soul. In other words, it is lack of knowledge of the true nature of the self which really constitutes moha; consequently, it is the knowledge (gnosis) and realisation of the self's true nature which banishes moha (aśuddhopayoga) and, by revealing and realising the inherent purity of the soul, accomplishes liberation. The role of material karman in this mechanism of bondage and liberation has thus for all significant purposes been forgotten. And it can be forgotten because the logic of the system no longer requires it. I refer to 'material' karman because its function has actually been taken over by immaterial aśuddhopayoga / moha, caused, in the niścaya view, by the jiva's self-transformation. As the Tattvadīpikā on Pravacanasāra 2.91 puts it:
Therefore the instrumental cause of the soul's connection with other substance is merely the absence of an accurate distinction between what is self and what is other.35
In other words, given that the cause of the jīva's connection with other substance is the delusion that the soul
appagam, two separate words, but how this is translated makes no difference to the present argument.
35 ato jīvasya paradravyapravsttinimittam svaparaparicchedābhāvamātram eva - TD on Pravac. 2:91.

Page #156

142 Harmless Souls is involved with the non-soul (including material karman), then the implication is that, since in reality the jīva has no connection with anything material, it cannot in reality be bound by karman. To believe otherwise amounts to moha, and it is this which is the 'fact' of the soul's bondage. Consequently, the belief that material karman can bind the non-material soul takes over karmapudgala's binding (i.e. its only) function, and Kundakunda has, in effect, internalised or dematerialised karman; bondage is now a matter of delusion, a false attitude manifested in a feeling of possessiveness with regard to the material (everything that is ajīva).36
The emphasis on moha, defined as the delusion that the self and the 'other' have any connection with or influence on each other, renders any 'two-cause' system of bondage - whatever the material and instrumental causes - effectively redundant. For how can something which does not and cannot really have any connection with what it is supposed in part to act upon be cited as a cause, material or otherwise? Furthermore, since it is the self's contact with paradravya, including material karman, that is a delusion, then it is paradravya / material karman that becomes, from the soteriological perspective, an irrelevance: in the characterisation of the Tattvadipikā on Pravacanasāra 2.101, 'unreal' (asat).37 In short, the stress on moha points to a solution to the problem of bondage beyond that attempted through the standard use of the vyavahāra / niscaya distinction. (The way in which moha arises, whether it is self-generated and how, or whether it is, like the Vedāntic avidyā, inexplicable, is not dealt with in this text; and for practical purposes it is probably considered
36 In this context, we may recall gāthā 1.46 (quoted p. 121, above): 'There would be no samsāra for any embodied jīva if the ātman itself, through its own nature, did not become auspicious or inauspicious.' Again this puts the 'blame' for bondage squarely on the ātman itself; the process is self-contained and internal.
37 For comments on this, see below.

Page #157

Kundakunda: The Pravacanasāra 143 irrelevant.)
In this way, Kundakunda has silently rid himself of an embarrassment, since, to make logical sense of bondage when karman and the jīva belong to two ontologically different categories (the material and the immaterial), he had two options: either to make the jīva material (and so revert to what may have been the very origins of Jaina doctrine) or to make karman immaterial. That he has, in effect, chosen the latter is borne out by evidence from both the Pravacanasāra itself and the Tattvadīpikā. Thus the Tattvadīpikā on Pravacanasāra 2.97,38 when it gives the niscaya view, actually refers to karman as though it were non-material: 'karman of/ for the ātman is modification into attachment'; 'it is duality of merit and demerit'. 39 In other words, karman - what binds - is, from this perspective, not something material (pudgala), but the modification of the immaterial self into states of consciousness (śubha- and aśubha-upayoga) such as rāga , which are meritorious or demeritorious. It is because the ātman is the agent (kartā), as it is also the appropriator (upādātā) and relinquisher (hātā), of modification into attachment, etc., that the latter is seen as its karman.40 This is stated explicitly by Kundakunda at Pravacanasāra 2.30, where it is said that:
The self itself is modification, and such modification is held to be action which consists of jīva; action is known as karman, therefore the self is not the agent (of material karman).
The Tattvadīpikā explains that, from the higher view, the ātman is the agent of that bhāva-karman which is in essence the modification of the self, but it is not the agent of dravya-karman which is in essence a modification of
38 Cf. Pravac. 2:29-30 and TD.
39 rāgapariņāma evātmanaḥ karma, sa eva punyapāpadvaitam - TD on Pravac. 2:97.
40 rāgādiparināmasyaivātmā kartā TD on Pravac. 2:97.

Page #158

144 Harmless Souls matter.41 Similarly, at Pravacanasāra 2.92, it is affirmed that while the soul is the agent of its own nature it is not the agent of all those states composed of material substances. Thus, as the Tattvadīpikā points out, the karman of the soul (taken here in the sense of action done by the soul, but therefore also its karmic matter, what binds it) is its own modification (svapariņāma). The soul does not bring about states of matter - it is not their agent - so they cannot be its karman. For the self, therefore, karman is not modification of matter.42 And Pravacanasāra 2.93 continues with the statement that, although at all times it exists in the middle of matter, the jiva neither grasps, releases, nor brings about material karmas.
These two gāthās (2.92-93)43 therefore offer a flat contradiction to the view (outlined above) that the soul is the instrumental cause of karman, and vice versa. That is to say, it contradicts the doctrines contained in Kundakunda's own niscaya / vyavahāra distinction. Dixit, who also notices this, identifies the 'thesis' put forward in these gāthās as Kundakunda's 'transition-point in his journey away from the traditional stand-point and towards the stand-point of the Samayasāra'.44 Given the apparently disparate nature of much of the Pravacanasāra, this is probably too specific an attribution; nevertheless, the tendency is clear.
Despite the views given at 2.92-93, the next three
41 tatas tasya paramārthād ātmā ātmapariņāmātmakasya bhāvakarmaņa eva kartā, na tu pudgalapariņāmātmakasya dravyakarmanaḥ - TD on Pravac. 2:30. The vyavahāra view is that it is modification of matter which is karman for the atman (pudgalapariņāmātmanaḥ karma, TD on Pravac. 2:97), and that the atman is the agent, appropriator and relinquisher of the modification of matter (pudgalapariņāmasyātmā kartā, TD on Pravac. 2:99; Cf. 2:29 and TD, quoted above).
42 evam ātmanah pudgalapariņāmo na karma - TD on Pravac. 2:92.
43 Cf. Pravac. 2:68-70. 44 Dixit 1971, p. 134.

Page #159

Kundakunda: The Pravacanasāra 145 gāthās (2.94-96) reiterate that sometimes and somehow (either because its space-points are stained by moha, rāga and dveşa [2.96], or because its association with rāga and dveşa brings about a self-modification into śubha and aśubha (-upayoga) [2.95]), it is clung to by karmic dust. As we have already seen, there is no satisfactory explanation of this conjunction, and the Tattvadīpikā on these passages comes no closer to providing one. For it states that the self, being its own agent, modifies itself, and the modifications of everything not self are only the occasioning causes of self-modification. How these modifications which are not self cause self-modification when the two have no contact is not explained. It is simply claimed that matter treats the soul's self-modification as an efficient cause to modify itself into karman,45 and that this karman then enters the self through the latter's (vibratory) activity. 46 (The perfunctory introduction of yoga here seems to be a sudden recourse to an earlier strand of doctrine.) Similarly, the soul treats matter as an efficient cause (nimitta) to transform itself into various states of consciousness.47 How this transformation takes place and how karman is actually bound remain unexplained features of this circular doctrine in which matter uses its proximity to the self-modifications of the soul to transform itself, and the soul uses its proximity to the self-modifications of matter to do likewise,
If I have laboured these problems, it is simply to make the point that it is the very attempt to preserve the soul's immaterial autonomy, and yet at the same time explain the fact of its bondage, which opens the door to a radically different emphasis in the theory of bondage and liberation. For it is precisely this struggle which gives rise to a perspective from which karman can - at least in some lights
45 tasya svapariņāmam nimittamātrīkstya - TD on 2:94. 46 yogadvārena, TD on Pravac. 2:95. 47 See TD on 2:97, and JPP p. 117ff.

Page #160

146 Harmless Souls - be viewed as immaterial.48 By stressing the niscaya view and the role of moha in the key passages on bondage in the Pravacanasāra, Kundakunda implicitly acknowledges the importance of a 'dematerialised' instrument of bondage for the development of his soteriology (a soteriology which, as we shall see, is taken to its logical conclusion in the Samayasāra). In this context, 'karman' as employed at Pravacanasāra 2.87 - 'Know this to be the niscaya description of the bondage of souls: the impassioned self binds karman, the self free from passion is free from karmas' - may therefore be taken to mean not material karman but the 'ātman's karman', i.e. the jiva's own manifestation of consciousness, (aśuddha-) upayoga.49
Again looking forward to the Samayasāra, and taking the niscaya view in the strictest sense of there being no contact at all between the omniscient (i.e. liberated) soul and matter, we read at Pravacanasāra 1.52 that:
Although knowing those objects, the soul does not transform itself ('under their influence' - Upadhye), does not grasp them, nor is it born among them; thus it is recognised as being without bondage.50
This gāthā describes the condition of the liberated or omniscient soul. However, as will become clear, in the Samayasāra, Kundakunda comes to view this kind of statement as the exclusive truth about the soul. In other words, it is not and has never been really bound, but has
48 See above, comments on Pravac. and TD 2:92, 2:97.
49 It should be noted that this reading also takes us closer to the way in which the naya doctrine is frequently used in the Samayasāra: there, the niscaya view is that the jiva has nothing whatsoever to do with paradravya, and both the elements - niscaya and vyavahāra - of the doctrine as it was introduced in the Pravacanasāra now fall into the vyavahāra camp.
50 Cf. Pravac. 1.32: 'The omniscient lord neither grasps, nor releases, nor transforms the other; he sees everything and knows everything completely.'

Page #161

Kundakunda: The Pravacanasāra 147 only been thought to be so through the supposition of an, in reality, non-existent relationship between jīva and ajīva.51 The corollary of this view - that the jīva binds itself through false consciousness or moha - is that it can liberate itself by attaining true or pure consciousness via meditational techniques which lead to the realisation of that fact. Since this false consciousness takes the form of a delusion that the self is in some way connected to the material world, when in fact it is intrinsically and always pure, then it is only a short step to the conclusion that the world of matter, and especially material karman, is less 'real' than the and even totally unreal. By the same token, it is possible to conclude that the soul was never really bound but only thought to be so. And these are in fact very close to positions taken by Kundakunda in the Samayasāra, although, as we have seen, there are also gāthās in the Pravacanasāra where they are strongly suggested, for instance, at 2.101.
It is worth looking briefly at Amộtacandra's commentary on this last passage (Pravacanasāra 2.101) as an example of the direction Jaina theory can take once karman has been effectively dematerialised. (And it is probable that the commentator uses his knowledge of the Samayasāra to provide himself with an interpretative framework here.) The text reads:
Bodies, goods, pleasures and sufferings, enemies and friends are not eternal for the soul (jīva). Eternal is the self (ātman) that consists of upayogu.
51 The ontological status of samsāra vis à vis the kevalin's omniscience cannot be entered into here; nor can there be consideration of what it can mean to 'know' something without having any relationship with it. However, the gist seems to be that the kevalin is in some sense co-extensive with samsāra - even the 'creator' of it - and so, in effect, the object of the kevalin's knowledge is itself alone, the self and what it knows being synonymous. See especially Book 1 of the Pravacanasāra, 1:23-1:36 (discussed below).

Page #162

148 Harmless Souls The Tattvadīpikā on this begins as follows:
For nothing is eternal for the self which is other than the self) and which is, since any such thing is not distinct from other substances and is indeed separated from its qualities as it undergoes the influence of other substances, the basis of impurity. [It cannot be eternal) because being unreal-and-caused (asaddhetumattvena) it must have a beginning and an end in time, and because it is not self-established (parataḥ siddhatvāt). 52
The introduction of the term asat, the 'non-existent' or 'unreal', to describe everything that is other than the self (i.e. everything that is paradravya or ajīva, including material karman) is radical, and yet, given the fact that the instrument of bondage has been internalised and the self established as absolutely different from and untouched by the 'other', not unpredictable. In other words, for soteriological purposes (and perhaps ontologically too) samsāra, made up of matter, including material karman, may be treated as irrelevant. (And we may also note here that karman and bondage by karman, being non-eternal, are, in contrast to orthodox Jaina doctrine, assumed to have a beginning.)
Soteriological interest is now focused entirely upon the realisation of the eternally pure ātman, which is the one way of destroying the bondage of moha. As the next gāthā (Pravacanasāra 2.102) puts it:
He who, knowing this, meditates with purified self on the paramātman (or 'the self and the other'], with formed or formless 'thought', destroys the tight knot of delusion.
52 ātmano hi paradravyāvibhāgena paradravyoparajyamānasvadharmavibhāgena cāśuddhatvanibandhanam na kimcanāpy anyad asaddhetumattvenādyantatvāt parataḥ siddhatvāc ca dhruvam asti - TD on Pravac. 2:101.
I owe the translation to Alexis Sanderson's interpretation of this passage and the TD. on Pravac. 2:100.

Page #163

Kundakunda: The Pravacanasāra 149
Since the effects of dematerialising karman and of seeing the soul as eternally pure are so far-reaching, it is little wonder that the Jainas struggled, by means of the naya doctrine, to preserve at some level the orthodox views that karman is material and that the soul has always been bound by it. For, liberated by this same naya distinction, Kundakunda is veering not just towards the heterodox, but towards the heretical; his doctrines entail not only a change in Jaina theory or 'philosophy', but, more importantly, they imply wide-ranging changes in Jaina practice, and so pose a threat to the whole social structure of Jainism as an independent Indian religion.
5.2 Moha
Before examining Kundakunda's mechanism of liberation in detail, it is useful to say something more about the role of what, according to parts of the Pravacanasāra, has become the effective instrument of bondage: namely, moha and its attendant 'passions'.
The classical position, outlined at Tattvārtha Sūtra 8:9, is that there are 28 kinds of deluding (mohaniya) karmas. Without enumerating all the subtypes, the basic division here is between 'perception-deluding' (darśana-mohanīya) karmas and 'conduct-deluding' (cāritra-mohaniya) karmas.
That is to say, mohanīya karmas produce moha or avidyā by deluding the jīva with regard to its belief and its conduct; or as P S. Jaini puts it, mohaniya-karman is karman that prevents the true perception of reality and the purity of the soul'. 53 In simplified diagrammatic form the division appears as follows:54
53 JPP p. 346 (Glossary); see also ibid. p. 131.
54 For a full enumeration, see SS on TS 8:9; Glasenapp pp. 8-11; JPP pp. 118-121.

Page #164

150 Harmless Souls
mohaniya, karman darśana-mohanīya-karman cāritra-mohaniya -karman engenders 3 degrees of
engenders the various passions, mithyātva karman
(kaṣāya), quasi-passions (nokaşāya) and their subtypes, the main division being:
rāga
dvesa
māyā lobha krodha māna
Jaini identifies these 'producers of delusion' as being at the heart of bondage, generating defilement which is characterised by confusion and desire and so causing the soul to become 'confused and desirous' 55 Thus, at one level, darśana-mohanīya and mithyātva karman clearly operate with regard to Jaina religious belief in general, and are explained as such. However, the same terms also carry a more specific weight. Tatia summarises this when he writes that in Jainism,
the term mithyātva (perversity) is generally used to denote the idea of avidyā. The terms mithyādarśana or mithyādęsti (wrong view), darśanamoha (delusion of vision), moha (delusion) etc. are also used in the same sense.
He goes on to say that the function of darśanamoha is to delude the soul and misguide it; under such influence the jīva accepts, among other things, the ajīva as the jīva.56 (Such a use of mithyādarśana can be found, for instance, at Tattvārtha Sūtra 8:1, Bhāşya and Sarvārthasiddhi.)57
55 JPP p. 117. 56 Tatia p. 144.
57 It should be remembered that Jaina philosophy uses the term darśana to denote both 'doctrine' or 'belief and 'undifferentiated

Page #165

Kundakunda: The Pravacanasāra 151 Thus, as P S. Jaini puts it, the mithyātva state, eigendered by darśana-mohanīya-karman, is manifested as a 'fundamental tendency to see things other than as they really are! 58
Cāritra-mohanīya-karman generates the various passions (kaşāya), and these two forms of mohanīyakarman combine to produce a 'condition of spiritual stupefaction' or defilement. 59 In this state the soul is open to the influence of other types of karman, affecting its remaining qualities and, indeed, generating its state of embodiment.60 Thus mohaniya-karman is of central importance to the classical theory of bondage; and although, as Jaini points out, it cannot be given first place in a beginningless cycle, it is impossible to eliminate other karmic influences 'as long as deluding factors remain'.61
However, although the importance of mohanīyakarman can hardly be denied in the classical theory, I should like to stress that it is the second term in the compound, 'karman', upon which the emphasis falls in terms of achieving liberation from such bondage. It is the fact that it is caused by deluding-(material)-karman that is of crucial significance, not the delusion itself, which is simply the particular form that karman takes. This becomes clearer if we look at the question of caritra
cognition' or 'perception' - see Glasenapp pp. 6-8.
58 JPP p. 118. 59 Ibid. p. 119. 60 Ibid. pp. 120-121.
61 Ibid. In this respect, mohanīya-karman has exactly analogous status to that held by avijjā ('ignorance and a synonym for moha) in the Buddhist formula of Dependent Origination (pațicca-samuppāda): aviijā stands first in the latter because it is the 'primary root of all evil and suffering in the world' [Buddhist Dictionary p. 31), and whereas it cannot be viewed as the 'causeless cause of the world (Visuddhi Magga XVII, 36f., quoted in ibid.), it is, figuratively speaking, a root-cause. In particular, it is the root-cause of lobha / rāga and dosa, 'consequently all unwholesome states of mind are inseparably bound up with it' (ibid. p. 32). Cf. the identical Jaina view given by the TD on Pravac. 2:103.

Page #166

152 Harmless Souls mohaniya-karman in slightly more detail, and thereby also remind ourselves of the essentials of Umāsvāti's kaṣāya doctrine.
In the classical theory, passion (kaşaya) is said to be generated by destructive karmas of the conduct-deluding (cāritra-mohaniya) kind. Passion is two-fold, consisting of aversion (dveşa) and attachment (rāga); 'the former is always divided into anger (krodha) and pride (māna) and the latter into deceitful manipulation (māyā) and greed (lobha)'62 It is the passions which produce pramāda ('carelessness') and thus himsā, i.e. the latter is the product of volitional activity motivated by passion.63 Thus kasāya, in effect, underlies all bondage: yoga attracts karmic matter to the jīva in the first place, deciding its type and quantity, while kasāya causes it to adhere, deciding its duration and intensity (Tattvārtha Sūtra 8:2-3). Passion is therefore necessary for bondage to take place. As we have seen, 64 parigraha (attachment to possessions) has the same nexus of associations as kasāya in general and rāga / lobha in particular. And at Tattvārtha Sūtra 7:17 parigraha is defined as mūrcchā ('infatuation' or 'delusion'), i.e. the delusion that something can be 'mine', as the Sarvārthasiddhi makes clear. Although, according to the Sarvārthasiddhi, infatuation or attachment (parigraha) 'is at the root of all evils' 65 - i e. it leads to himsā, etc. - passion, nevertheless, has the controlling hand, and remains instrumental. To quote the same source, 'so the passionless person posses ed of right faith, knowledge and conduct is free from delusion (moha). Hence there is no infatuation in his case'. 66 So even when the term 'moha' is used in the
62 JPP p. 119; see TS 6:5 (6), cf. 8:1. 63 See above, pp. 54-55. 64 See above, p. 31ff. 65 tanmūlāḥ sarve doşāḥ - SS on TS 7:17.
66 tato jñānadarśanacāritravato 'pramattasya mohābhāvān na mūrccha 'stīti nisparigrahatvam siddham - ibid., trans. by S.A. Jain p. 199.

Page #167

Kundakunda: The Pravacanasāra 153 Sarvārthasiddhi, it designates something more affective than cognitive in character, contrasting with its employment in Kundakunda as something essentially cognitive.
The consequence of viewing kasāya as the instrumental cause in bondage is that the mechanism of liberation in the Tattvārtha Sūtra is directed towards the individual ridding himself of such passions, and thus both stopping the influx of, as well as shedding, the karmas which are caused to be bound by passions. Tapas (austerity) 'effects both (stoppage and dissociation) and ... is the chief cause of stoppage of influx'. 67
Tapas is divided into 'external' and 'internal' austerities (see Tattvārtha Sūtra 9:19-20). Both categories are predominantly physical in character, although meditation (dhyāna) is included as the last of the internal austerities. Such meditation, however, is defined at Tattvārtha Sūtra 9:27 as 'concentration of thought on one particular object'. 68 It is not designed to engender gnosis or realisation of the true nature of self - in contrast to what we find in Kundakunda, where it is effectively moha, and not kaṣāya as such, which is seen as the root cause of bondage (see above). • In the Tattvārtha Sūtra, therefore, the instrumental causes of bondage are basically twofold (or, as subdivided, fourfold): rāga and dveşa (māyā and lobha, krodha and māna). In the Pravacanasāra, however, rāga is used - perhaps in a more general sense - to denote 'passion' or 'attachment', and is usually part of a threefold division of the causes of bondage, completed by dveşa and moha. Rāga and dveșa are self-evidently a pair, or two sides of the same emotion; the employment of one towards one 'thing' entails the employment of the other towards another thing'.
67 ubhayasādhanatva ... samvaram prati prādhānya - SS on TS 9:3, trans. S.A. Jain p. 242.
68 ekāgracintānirodha - TS 9:27 trans. S.A. Jain, p. 266.

Page #168

154 Harmless Souls And as we have seen,69 when rāga is associated specifically with parigraha, and dveşa with ārambha, then the former leads to or causes the latter. In the Pravacanasāra and Tattvadipikā, moha, the newly introduced third element, is seen as being the root-cause of the other two. 70
Why does Kundakunda add moha to the standard twofold kaşāya doctrine? There are probably two related reasons: his dematerialisation of the cause of bondage (karman) (outlined above), and the influence of the standard Buddhist triad of karmically unwholesome roots (mūla), viz. greed (lobha = rāga = tanhā), hate (dosa), and delusion (moha). It hardly needs pointing out that in a system where the jiva 'really' has no connection with matter, and where karmic bondage is seen as being essentially cognitive in nature, delusion or its synonym, 'ignorance', is likely to be suggested as the root cause of that bondage. In other words, even without the Buddhist model, some such grouping of rāga and dveșa with moha was likely to have suggested itself to Kundakunda, given his stress on the total separation of soul and matter and his movement towards an immaterial karman. Moha thus has a similar role in Kundakunda's teaching to avidyā in Śankara's Advaita Vedānta and avijjā / moha in the Pāli Canon.
In the Pravacanasāra, therefore, we are presented with what appears to be a less complex formulation than that given in the Tattvārtha Sūtra: moha is added to rāga and dveşa and the three together constitute or give rise to aśuddhopayoga, and are thus responsible for bondage. However, Kundakunda's formula is clearly a refinement of the classical position in terms of his upayoga doctrine, and should not be supposed to predate it. His position appears to be less complex because the prolix categorisations of the
69 See above, Part I passim.
70 See above, pp. 141-142. As in the TS it is simply mohaniyakarman which is the generating cause of rāga and dveșa.

Page #169

Kundakunda: The Pravacanasāra 155 underlying material karman theory have been pushed into the background or 'forgotten'. In other words, there are few indicators that every time one sees 'moha' one should understand 'material mohaniya-karman'; on the contrary, moha as used by Kundakunda has clear non-material associations, and in this respect the fact that the Jaina's formulation reproduces the Buddhist triad of 'unwholesome roots' must be the result of sympathy rather than the attempt to 'colonise' a rival doctrine through re-definition. Some of the coincidences between the two formulations have been outlined above. But in summary I would suggest that Kundakunda's employment of 'moha / avidyā' is closer to the Buddhist use of the term, with its cognitive and volitional associations, than it is to the material mohanīyakarman of standard Jaina doctrine, with its ramifications for ascetic practice (i.e. its requirement of extreme forms of tapas).71
The fact that Kundakunda uses mohaniya-karman as a category at least once in the Pravacanasāra,72 but prefers to concentrate on the less obviously material moha, shows both his awareness of traditional doctrine and where he chooses to lay his emphasis. Faddegon, commenting on the use of duttho / dusta (glossed by the Tattvadīpikā as moha) at Pravacanasāra 1:43, remarks that moha, rāga, and dosa | dveșa together form mohaniya-karman, and contrasts this triple division with 'the more intricate classification of mohanīya-karman' in the classical position (see, for example Tattvārtha Sūtra 8:9).73 While not wishing to
71 It should be noted that in Jayasena's recension of the Pravacanasāra there is an extra gāthī at 3:26b where the four passions are referred to in a perfectly orthodox, i.e. classical, sense. This is an exceptional case, and it is perhaps significant that it occurs in Chapter 3, which appears to be only loosely related to the first two chapters and may well have been compiled for a different purpose.
72 At 1:15 and TD; cf. 1:19 and 1:43.
73 Faddegon [Kundakunda (3)] p. 27, fn. 1. The full text of Pravac. 1:43 reads: 'The great Jinas say that portions of karmas are necessarily operating (and giving their fruit); he, who is infatuated with,

Page #170

156 Harmless Souls deny that Kundakunda's idea of moha ultimately derives from and sometimes reverts to the classical position, I would point to its role in a new formulation, its association with the upayoga doctrine, and especially its connection with realisation (jñāna) and not tapas, as indicators that the emphasis in the Pravacanasāra is on an effectively dematerialised instrument of bondage. The complexity of the orthodox classification of (physical) mohaniya-karman has thus been largely by-passed simply because an understanding of it is essentially irrelevant to the proposed means of liberation (jñāna through dhyāna); whereas, for reasons outlined above, the orthodox classification points towards samvara and nirjarā through tapas as the means to liberation.
5.3 Himsā in the Pravacanasāra
i) The role of himsā in the Pravacanasāra With the stress on moha, its corollaries' (avidyā and aśuddhopayoga) and antidotes (jñāna through meditation on the essential purity of the self, and suddhopayoga) the characteristic Jaina ptoccupation with the ethical imperative of ahimsā would appear, in the Pravacanasāra, to have been moved to the periphery of soteriological concern. How then does Kundakunda view himsā, and what, according to him, is the relation of the latter to, for instance, moha?
In Book 3 of the Pravacanasāra we find a group of gāthās dealing with himsā which, taken without the commentary, seem to have only a tenuous connection with the kind of doctrines expounded in the first two books. These verses certainly look more traditional and more orthodox; they are closer in feeling to the Tattvārtha Sūtra than to the Samayasāra. Nevertheless, they merit more
or shows attachment or aversion towards them, necessarily incurs bondage.

Page #171

Kundakunda: The Pravacanasāra 157 detailed examination, for they provide some indicators of major changes in Jaina religious practice, and help to clarify the even greater changes implicit in Kundakunda's less orthodox doctrines.
At Pravacanasāra 3:16 himsā is defined (in Upadhye's translation) as follows:
Careless activities of a monk when sleeping, sitting, standing and walking, are always known as continuous harm unto living beings.
This would seem to be the most natural rendering, i.e. himsā is harm to living beings as a result of careless physical actions on the part of the monk. Prima facie, this is a purely orthodox doctrine such as might be found in the Dasaveyāliya Sutta. Amrtacandra, however, interprets it in the Tattvadipikā in terms of Kundakunda's upayoga doctrine, an interpretation reflected in Faddegon's translation: 'Heedless action in lying, sitting, standing, going, etc., is considered to be at all times continual hurt (himsā) to the śramana-state!.
Thus Amstacandra (and Faddegon) take the Prākrit samaņassa to be equivalent to the Sanskrit śrāmanyasya, rather than śramaṇasya, i.e. they add the suffix -ya- to śramana and so turn it into an abstract meaning 'śramanastate'. 74 There seems to be no linguistic justification for this, but, as we shall see, it is resonant in terms of doctrinal change. (And although the original Prākrit gāthā may not be by Kundakunda, the interpretation given in the commentary demonstrates the kind of practice his upayoga doctrine implies, and is thus highly relevant to the present enquiry.) The Tattvadīpikā is worth quoting in full here. Amrtacandra's introduction to the Prākrit gāthā tells us that this is a definition of what constitutes cheda - 'infringement' or, more specifically, an offence against the monastic rules. The commentary proper states:
74 Cf. Pravac. 2:98 where samannam = śrāmanyam.

Page #172

158 Harmless Souls
For negligence is impure manifestation of consciousness because of the destruction, the injuring, of that śramaņa-state whose form is pure manifestation of consciousness. And so (negligence) itself is himsā. Therefore, the śramana's careless activity with regard to sleeping, sitting, standing and walking, etc., which is inherently connected with impure manifestation of consciousness, is always himsā for him, a continuous force (lit. flowing continuously) at all times, which is another disadvantage consisting of infringement.75
Interpreting this, we may say that the monk 'hurts' himself - i.e. binds himself - through aśuddhopayoga of which cheda is the external sign. For since the ideal śramaņa-state has the form of suddhopayoga, then aśuddhopayoga is, in the first place, himsā to the śramana who aspires to that state, and, in the second, it necessarily entails (physical) cheda. That is to say, offence against the monastic rules presupposes aśuddhopayoga. So the original or initiating cause of himsā, and thus of bondage, is internalised: the emphasis is on aśuddhopayoga, not on external cheda as such. Moreover, himsā has been (re-) defined as an offence against the śramaņa-state, i.e. an offence against the monastic rules (= cheda), and thus as harm to the self because of the aśuddhopayoga that is necessarily at the root of such neglect, rather than harm caused to other living beings as such. So it becomes apparent that it is now the role of the śramaņa which is all important; obedience to the prescribed actions not only leads automatically to the goal, but, more importantly, it does so only because - as that very obedience indicates (externally) - internal purity (śuddhopayoga) has been maintained, for there can be no adherence to the monastic rules without such purity. Consequently, both monastic
75 aśuddhopayogo hi chedaḥ śuddhopayogarūpasya śrāmaṇyasya chedanāt tasya himsanāt sa eva ca himsāataḥ śramaṇasyāśuddhopayogāvinābhāvini sayanāsanasthānacankramaņādişv aprayatā yā caryā sā khalu tasya sarvakälam eva samtānavāhini chedānarthāntarabhūtā himsaiva || TD on Pravac. 3:16.

Page #173

Kundakunda: The Pravacanasāra 159 behaviour and ahimsā may be said to have been ritualised in the sense that it is now the mechanical performance of the prescribed action itself that counts, or is indicative, not the ahimsic content of the action.76 What is significant is the harm caused to oneself through aśuddhopayoga rather than any harm that might be caused to others, although this entails no particular change in practice since the monastic rules, adherence to which is now the expression of innerpurity, were originally formulated to prevent harm to others. (That is to say, while originally it was harm to others that caused harm (= bondage) to the self, it is now aśuddhopayoga, a state of consciousness, whose external expression is cheda, offence against the monastic rules, that does so.)
In other words, obedience to the rules leads automatically to the goal, but only in the sense that such obedience refers back to, or is a necessary consequence of, an internal state of suddhopayoga, which is what really counts for salvation. Thus total externalisation (the ritualisation of ahimsā) and total internalisation meet: one implies, or is a reflex of, the other. And although, theoretically, it is the internal (suddhopayoga), as the agent of the external (monastic practice), which is really instrumental in binding and liberating, in terms of the visible - of external practice - it is simple adherence to the letter of the law which is seen to bring about liberation. But, again, it is internalisation of himsā that allows it to do SO.
As we have seen, it is the commentary and not Kundakunda's text (3:16) which engendered the above analysis. However, the following gāthā is more directly relevant to at least part of that interpretation:
Whether the being dies or lives, injury is certain for the man who is unrestrained; there is no bondage simply by injury (i.e. 'by mere
76 On my use of the term 'ritualisation' (the meaning is close to routinisation or formalisation), see above p. 86.

Page #174

160 Harmless Souls
physical harm' - Upadhye) for the man who regulates himself according to the rules of conduct.
This gāthā has already been discussed above, 77 since it was quoted without attribution by Pūjyapāda (Devanandin) in the Sarvārthasiddhi on Tattvārtha Sūtra 7:13. It is possible that Pujyapāda was referring directly to this gāthā from the Pravacanasāra (as the Jaina Granthamālā edition of the Sarvārthasiddhi claims), but it is more likely that both 'authors' quote from some earlier, unidentified, (possibly) canonical source. And in any case, as we have already noticed, this part of the Pravacanasāra shows every sign of having been compiled from other sources rather than composed by Kundakunda himself.78
For the purposes of the present enquiry, the important thing is to examine the way this section is brought into line by the commentary with the doctrines propagated by Kundakunda in Books 1 and 2 of the Pravacanasāra. In this context, therefore, gāthā 3:17 means that as long as the śramaņa retains his internal purity, reflected externally in his adherence to the letter of the monastic rules, then he cannot do himsā or be subject to its effects (i.e. be bound), even if physical harm is caused. (Note the intentional ambiguity of himsā: it is injury done to others and therefore also to oneself.) In other words, outside the parameters set by the monastic code - which is the external reflection of internal purity - physical action and its consequences are irrelevant; it has no karmic and, therefore, no soteriological significance.
Amrtacandra introduces this gāthā (3:17) with the claim that in it Kundakunda teaches two kinds of cheda, antaranga (internal) and bahiranga (external). He defines them as follows: 'Impure upayoga is internal infringement,
77 See above, p. 53ff.
78 Upadhye (p. liiiff.) comments that this verse has 'a traditional appearance, and [it] might be traced to a tract of literature which was once the common property of Digambaras and Svetāmbaras'.

Page #175

Kundakunda: The Pravacanasāra 161 the taking of another('s) life is external'.79 Whether it happens that another being's life is taken or not, the fact of 'careless conduct' (aprayatācāra) proves aśuddhopayoga (i.e. the external is a sign of the internal state), and so the existence of himsā for that impure consciousness is certain.80 (Read with 3:16 above, one may also understand from this that aśuddhopayoga causes himsā to the self.) Conversely, 'careful conduct' (prayatācāra) proves the nonexistence (asadbhāvasya) of aśuddhopayoga (internally); so himsā does not occur for that consciousness, as is shown by the fact that there is no bondage (for that pure consciousness) even as the result of taking another's life.81 Therefore, Amstacandra concludes, internal cheda is more powerful or important (balīyaḥ) than external. Nevertheless, external infringement should be admitted simply because it provides a place (or "seat") for internal infringement'.82 In other words, external cheda is merely a sign of, or an emblem for, an internal state; the latter manifests itself in the former.
This is repeated and amplified in the Tattvadipikā on 3:18. The gāthā reads:
A Śramaņa of careless conduct is called a murderer of the six (classes of) embodied beings; if he carefully practises (his course of conduct) he is forever uncontaminated like the lotus on water.83
Again it is the performance to the letter of the monastic rules which is crucial here. To this ritualisation the Tattvadīpikā adds the internalisation which may also be
79 Trans. after Faddegon [Kundakunda (3)] of aśuddhopayoga 'ntarangachedah paraprāņavyaparopo bahirangaḥ.
80 suniścitahimsābhāvaprasiddheḥ - TD on Pravac. 3:17. 81 paraprāņavyaparopasadbhāve 'pi bandhāprasiddhyā - ibid.
82 evam apy antarangachedāyatanamätratvād bahirangachedo 'bhyupagamyetaiva - ibid.
83 Upadhye's trans. The six classes of living beings are the five sthāvara and the one trasa, i.e. the 'immobile' and the 'mobile'.

Page #176

162 Harmless Souls
latent in Kundakunda's gāthā. Carefulness is something interior, an attitude; conversely, it is aśuddhopayoga which is at the root of all external 'carelessness' and the initiating cause of bondage. That is to say, it is aśuddhopayoga, the cause of the violence, which entails bondage; himsă in itself is karmically neutral. Amṛtacandra works hard to stitch all this together:
Since the existence of impure manifestation of consciousness, proved by careless behaviour, which does not occur without it, is hurtful since it is established that bondage is caused by the destruction of the lives of the six kinds of bodies; and since the absence of impure manifestation of consciousness proved by careful behaviour, which occurs when it is absent, is not hurtful; and because of the absence of even a small amount of bondage caused by the 'other' (viz. the ajīva), so that it is established that it is free from impurity, like a lotus lapped by water; therefore it is concluded that internal infringement, which has the form of impure manifestation of consciousness, should be prevented by all means by which external infringement, in the form of taking another life, which merely provides the occasion for that (internal infringement), is a fortiori prevented.84
Superimposed in this one gathā and its commentary there may be as many as three layers of doctrine: 1) an archaic himsă doctrine where any - even accidental - harm is binding, 2) a ritualised doctrine, where what counts is
84 Trans. after Faddegon [Kundakunda (3)] p. 164, of TD on
Pravac. 3:18:
ca
tadvinābhāvinā
yatas tadavinābhāvinā aprayatācāratvenaprasiddhyad aśuddhopayogasadbhavaḥ ṣaṭkāyaprāṇavyaparopapratyayabandhaprasiddhyā himsaka eva syāt | yataś prayatācāratvena prasiddhyad aäuddhopayogāsadbhāvaḥ parapratyaya bandhaleśasyāpy abhāvāj jaladurlalitam kamalam iva nirupalepatva prasiddher ahimsaka eva syāt | tatastaistaiḥ sarvaiḥ prakārair aśuddhopayogarūpo 'ntarangacchedaḥ pratiṣedhyo yairyais tadāyatanamātrabhutaḥ paraprāṇavyaparoparūpo bahirangacchedo dūrād eva pratiṣiddhaḥ syāt ||

Page #177

Kundakunda: The Pravacanasāra 163 that the śramana should perform his prescribed role to the letter, and 3) an internalised hiņsā doctrine, linked to Kundakunda's upayoga theory, whereby the external is merely a reflex of the internal state, and it is the latter which carries the soteriological weight.
In other words, the physical model of the killing of the six classes of embodied beings is retained, but the original doctrine is now interpreted in terms of 1) ritualised action, and 2) internalised doctrine. Perhaps that is why the gāthā says vadhakaro tti mado, 'he is regarded as a murderer', i.e. this is just a way of talking about the person who is internally careless (which is what really counts), using the old imagery, the old physical connection of carelessness and jivas, in a new context. (Here, as elsewhere, ritualisation and internalisation complement each other.)
To expand on this, originally the five samiti can only have been formulated as rules for the avoidance of himsā to the various kinds of jivas.85 The emphasis of the discipline is to avoid doing harm because such harm would automatically cause one to be bound by a new influx of karman. Yet here the discipline of samiti has been ritualised; it is adherence to the letter of monastic discipline that is crucial; if a monk follows the rules then, even if he does harm 'accidentally' - which would count as 'carelessness' in the original reading -, he is not bound by karman. Emphasis is switched from the results of actions to the actions themselves, and thus to the underlying attitude or volition accompanying them. This combination of ritualisation and internalisation disposes of the worry that even someone who is careful and observes the five samiti can be open to 'accident'. In other words, it removes the fortuitous from monastic life, for now it is attitude, externalised in particular ritualised actions, that really counts. Yet, once that shift of emphasis has been made, the doctrine that himsā to jīvas (including nigoda) is what causes karmic bondage is undermined, for gāthās such as
85 See Part I, above.

Page #178

164 Harmless Souls Pravacanasāra 3:17 claim that 'unintentional' himsā - or himsā occurring after a certain degree of carefulness, encapsulated in the rules for ascetics, has been observed - does not bind. And if there is one case where himsā does not bind, then the logic of the system is cracked and has to be reassembled. For now it is something mechanical, a rule, that decides whether karman is bound or not, not actual himsā; at the same time, himsā is internalised and redefined to mean an attitude of carelessness, externalised in offences against the monastic rules. (To borrow a term from T.S. Eliot and the critical vocabulary of English studies, it might be said that offence against the rules is the 'objective correlative' of the subjective state or attitude.) Consequently, the karmic quality of an act is, in effect, decided, as in the Buddhist case, by the intention and attention - the internal state - which accompanies it, not by the act itself or what it results in. However, since that internal state is reflected or externalised so closely in the external rule, there is little visible alteration in behaviour. In this respect, the particular rationales that the Buddhists and the Jainas give for their differing external practices are each a mirror-image, or reversal, of the other. For the Buddhists, the purpose of rules of conduct (vinaya) is to help the monk to cultivate certain inner states which are instrumental in liberation. For the Jainas (as represented by Amộtacandra), however, the inner state has precedence over and conditions the outer: perfect conduct is not so much a means to the end as a reflection of the already internallyachieved end. In other words, for the latter, the rationale for external practice is not so much soteriological as social - a public demonstration of inner purity.
And in practice, this kind of public performance became important for Buddhism too. As Richard Gombrich points out, according to the view of ethics found in the Vinaya and commentarial literature (which was, of course, formulated with the Sangha in mind), sila (morality) is 'the monk's successful role performance. It makes clear to society he is being a good and proper

Page #179

Kundakunda: The Pravacanasāra 165 monk'.86 This external decorum was the response to 'an overwhelming demand for empirical evidence of a monk's internal state'.87 Thus the monk's moral habit, which is primarily an internal state, becomes 'instantly recognizable by a pattern of behaviour'.88
Such an explanation for the function of external practice is, of course, highly theoretical, since to achieve the optimum inner purity it is necessary to become an ascetic, and the only way to become an ascetic is to adopt the ascetic's vows. It is clear, therefore, that in practice the external must precede the internal. (The rationale works both ways: since the ascetic cannot follow his course of discipline without the necessary inner purity, the very fact that he is following it demonstrates to himself and others that he does have the requisite inner resources.) Nevertheless, the need for this type of theoretical justification of ascetic discipline demonstrates the extent to which the level of internalisation reached by Kundakunda and consolidated by his commentators offers a serious threat to standard Jaina practice. In other words, although the rationale for continuing ascetic practice may be weak, it is recognised that there does have to be some kind of rationale, otherwise the very complex of behaviour which provides the Jainas with their social and religious identity is made redundant. Thus, from this perspective, the real tasks of the post-canonical writers and scholastics are seen to be, on the one hand, the acknowledgement of the practical limits of 'carefulness' and, on the other, the need to rein in the logic of internalisation before it bolts and leaves behind any necessity for external practice, thus discarding the social identity such practice carries with it. As we have seen, a central element in this struggle to reinterpret ancient ascetic practices in doctrinal terms which are compatible with new social circumstances is the redefinition of what
86 Gombrich 1984, p. 100.
87 Ibid. 88 Ibid.

Page #180

166 Harmless Souls counts as himsā, in the sense of 'bondage-causing himsā'. For the whole of ascetic discipline is built on the theory that there are ubiquitous jīvas, that harm done to them constitutes himsā, and that himsā binds karman. In this context, the only way of setting practical limits to what counts as himsā (i.e. to 'carelessness') - that is, the only way to alleviate the formidable difficulties of ascetic practice and thereby give hope of liberation (or at least a better rebirth) to those who are not advanced ascetics - is to internalise it. Only after such a process is it possible to 'return', as it were, to the external world with a modified (i.e. ritualised) form of ascetic practice - limited to literal adherence to the monastic code - which can be defended doctrinally.
This combination of mental discipline and adherence to the letter of particular physical vrata (whether monastic or lay) may be considered, at least in one respect, less taxing than the ancient, blanket adherence to physical ahimsā, and also, perhaps, easier to practise from within, or on the fringes of, society. For, when external behaviour becomes formalised, chance is removed from Jaina soteriology. The external world becomes less threatening, and life less contingent for the ascetic, who is now in more or less total control of his progress to liberation. This movement towards absolute personal control was probably inevitable, given that the world portrayed in the earliest canonical texts was a risk-saturated environment, even for the ascetic. The early 'canonical' śramana is under constant threat from other people and from the physical world: they threaten him not just to the extent that he threatens them, but he is also constantly at risk from the fortuitous and the accidental.89 (That is to say, he is under constant threat of being forced into, or finding himself by circumstances beyond his control, in a position where ahimsā is inevitable; and so he is under continual threat of further bondage.)
Such 'accidents' could, of course, be rationalised in
89 See, for instance, the Āyāramga and Das.aveyāliya Suttas.

Page #181

Kundakunda: The Pravacanasāra
167
terms of karma theory, as the karmic fruits of some previous action. That kind of explanation, however, does not solve the problem of control. The ascetic develops selfdiscipline precisely in order to have absolute control over his karmic destiny; and absolute control is only possible in isolation from the contingent, material world. Consequently, the earliest Jaina discipline demanded almost superhuman control and concentration, a fact which must have severely limited the number of recruits. To survive, even as a śramana movement, it must sooner or later have been necessary, among other things, for Jainism to hold out the prospect of a sure path towards release, free from chance and accident, the equivalent of the Buddhist bhikkhu's being 'an island to himself.9
90
Kundakunda is at the logical end of this movement towards ever-increasing autonomy. In contrast to the position of the early 'canonical' śramaṇa, the greatest threat to Kundakunda's ascetic comes from within himself. Essentially, this threat amounts to ignorance of his own nature. Once that ignorance is dispelled, then the rest (i.e. proper external action) falls into place. If you have realised your true nature, then karmic bondage is impossible: whatever you do is himsã-free because it is the purified, knowing self that is doing it. No accident is possible where everything is jñāna - defined by Kundakunda as 'self' (Pravacanasara 2:67, etc.). To put it another way, there is no longer any real interaction between the ascetic and the world of matter, since the latter can have no real effect on the former as long as his attitude is correct. The contingency which characterises ordinary life is blocked out, and the ascetic remains safe inside his own consciousness, his own upayoga. Upayoga is the fail-safe mechanism which ensures that the ascetic's salvation is in his own hands, and through it he becomes autonomous.
This divorce of the śramana from the world of matter
90 See Digha-Nikāya II (Mahāparinibbāna-Sutta), quoted by Rahula p. 60.

Page #182

168
Harmless Souls
goes in tandem with a doctrine such as Kundakunda's (in the Samayasara and parts of the Pravacanasāra), which points towards the unreality of all except the pure soul. The śramana's internal state has thus become its own universe, hermetic, and for all practical purposes, idealist. He no longer struggles with the world of matter, but with himself, alone. At least, from the point of view of personal soteriology, this is probably the position in which Kundakunda and his followers would like to find themselves. However, since purity of attitude is impossible to demonstrate except through purity of action, in a social situation ascetic behaviour is the only indication of a holy life: monks and nuns remain exemplars in their behaviour. Such behaviour, therefore, is a means of mutually maintained social regulation: the ascetics do what the laity expect them to do, while in return, the laity remain faithful and support them while maintaining their own lower-grade but related practice.
ii) Himsă, moha, and upadhi
In Jayasena's recension of the Prayacanasāra, gāthā 3:17 is followed by two extra gāthās not contained in Amṛtacandra's version (3:17b and 3:17c). The fact that these two (as well as 3:17) are quoted (with minor linguistic differences) in the same part of the Sarvarthasiddhi (on Tattvärtha Sūtra 7:13) lends weight to Upadhye's suggestion that we are dealing here with verses that form a group in some other source. .91 The gāthās read:
When the foot of an ascetic who observes the irya-samiti (i.e. who is careful in his walking according to the rule) has been raised for going out, should a minute creature (kulimgam),92 coming in contact with that, be hurt or killed, it is taught in the scripture that he is not liable even for the slightest bondage as a consequence of that; (the case is similar to the statement:) it is infatuation alone
91
See note 78, above.
92 Kulimgam is glossed by Jayasena as sūkṣma-jantuḥ.

Page #183

Kundakunda: The Pravacanasāra 169 that is called possession on the authority of the spiritual lore. [Pravacanasāra 3:17b & 3:17c]93
In terms of doctrine these gāthās are probably closer to the Tattvārtha Sūtra, and its concern with kaşāya, than to Kundakunda's upayoga doctrine; i.e. prima facie, internalisation of himsā takes a different theoretical channel here. However, the redefinition of himsā by the collation of the elements of the first and the fifth mahāvratas (ahimsā and aparigraha),94 and the further equation of parigraha and mūrcchā, brings us back to moha as the agent of bondage. Thus it is infatuation (mūrcchā / moha), stemming from a mistake about the jīva's relationship to matter, which causes adherence to physical objects (parigraha), and so physical or 'external' himsā. "Internal' himsā, harm to oneself, is the mental state of 'infatuation', external harm merely the indicator of that internal state, and
93 Cf. Upadhye's trans. On iriyāsamida, Sk. iriyāsamiti, see TS 9:5 and SS; J.L. Jaini 1940, p.134; Schubring 1962, para. 173. Faddegon [Kundakunda (3), p. 201) gives the second half-verse of 3:17c a very different translation, viz. '... just as acceptance of swooning also is regarded according (as it is due) to (concentration) on the self. That is to say, swooning, and thus causing himsā by falling, has no binding effect when it is the result of meditation practice. However, although it is possible for mucchāpariggaho (Sanskrit: mūrcchāparigraha) to mean 'acceptance of swooning', in this context the interpretation seems strained. Given the close association of mūrcchā and parigraha in Jaina theory (see TS 7:17 and SS, where they are defined as equivalent), Upadhye's reading appears to be the correct one here. Moreover, the stress in these gāthās is clearly upon what Upadhye calls the 'mental condition', referring to which, his translation contains a footnote that is worth quoting in full:
'Himsā is not merely prāņa-vyaparopaņa, but pramatta-yogāt prāņa-vyaparopanam (TS. VII, 13). It is passions, negligent and careless channels of activities etc., that matter most; it is the mental condition, rather than the visible act, that is of utmost importance. For instance, parigraha does not so much consist in having physical contact with external objects as being infatuated with them' (Upadhye fn. 1, p. 26 on 3:17*1-2).
94 See above.

Page #184

170
Harmless Souls
it is the latter which is really instrumental in terms of binding and liberating the jiva.
This is part of the same pattern of theory as Pravacanasara 2:24, which remarks that 'there is no action without fruit, although the highest dharma is without fruit'.95 The Tattvadīpikā reads this in terms of 'mental' action, the action of one who thinks (cetana), and such action is defined as a modification (pariņāma) of consciousness (caitanya). However, this modification is only (karmically) fruitful (i.e. binding) for the atman connected with 'delusion' (mohasamvalita; sval(I) + sam). When the connection of the soul with moha disappears, the action is without fruit, so there is no further rebirth. And it is this fruitless (i.e. moha-less) mental action which Amṛtacandra defines as the parama-dharma, 'the highest dharma'. This may be compared with Pravacanasāra 2:58, where it is the idea of 'mine' (mamattam) with regard to external objects, especially the body, which is the cause of rebirth. (According to the Tattvadīpikā, this idea is the 'interior cause' of rebirth, a manifestation of attachment [uparaktatva].) Taking these two together, it is clear that it is the 'delusion of possession' - the idea that the jiva can have a real connection with anything ajīva - that is the real cause of rebirth. In other words, himsā, the binding instrument, has effectively been internalised to moha, which, in turn, is a manifestation of, or equivalent to, 'false consciousness' or aśuddhopayoga.
Pravacanasara 3:19 and commentary assemble a further set of equivalences around the concept of himsā which point to aśuddhopayoga as the significant cause of bondage. Gāthā 3:19 reads:
There is or there is not bondage, when a being dies in the course of physical activities; bondage is certain from attachment to
95 kiriyā hi natthi aphala dhammo jadi nipphalo paramo - Pravac.
2:24.

Page #185

Kundakunda: The Pravacanasāra 171 possessions, therefore ascetics should give up everything. 96
Upadhye renders uvadhi (Sk. upadhi) as 'attachment to paraphernalia', although he has already used 'paraphernalia' (i.e. physical possessions) to translate the term pariggaho / parigraha at Pravacanasāra 3:15. At 3:19 he follows Jayasena, who, in the Tātparya-vrtti, paraphrases upadhi by parigraha.97
Faddegon, in a footnote on 3:15, remarks that:
Tyāga is renunciation of all wordly concerns (parigraha), i.e. of possessions; vyutsarga is abandonment of all that may become seductive (upadhi), specially that which is required for the welfare of the body. Thus tyāga and vyutsarga, parigraha and upadhi are nearly identical.98
However, in common with a number of other technical terms, such as cheda, himsā, etc., upadhi can be shown to have two facets or areas of reference, one facing or referring to the external, physical world, and the other referring to an internal state or attitude. The earlier, technical sense (which is probably the one used at Pravacanasāra 3:15)99 is found, for instance, at Uttarajjhayana 29:34, where 'the renunciation of articles of use' (upadhi-pratyākhyāna) is prescribed. 100 The later internalisation of upadhi is evident in gāthās such as Pravacanasāra 3:19 (particularly when taken in conjunction with the next gāthā, 3:20), a phenomenon spelt
96 Based on Upadhye's translation.
97 As Faddegon (Kundakunda (3)] points out, p. 164, fn. 2. TV on Pravac. 3:19 reads: atha bahirangajivaghāte bandho bhavati na bhavati vā parigrahe sati niyamena bhavitīti pratipădyati.
98 lbid. p. 162, fn. 2. 99 Cf. Pravac. 3:23, where it has the same meaning.
100 Jacobi translates: 'By renouncing articles of use (except such as are obligatory - the broom etc.) he obtains successful study; without articles of use he becomes exempt from desires, and does not suffer misery' (1895, p. 167).

Page #186

172
Harmless Souls
out in detail in the Tattvadīpikā, where the connection of upadhi and aśuddhopayoga is made explicit.
In his introduction to Pravacanasāra 3:19 Amṛtacandra states that, 'Now he (i.e. Kundakunda) teaches that similarly attachment to possessions (upadhi), since it is. unequivocally an internal infringement, should be prevented.'101 He goes on to explain in the commentary proper that,
Whereas in the taking of another being's life through a bodily action the degree of infringement is held to be indeterminate (chedatvam-anaikāntika), - because its character as bondage is variable according to the presence or absence of impure manifestation of consciousness (aśuddhopayogasadbhāvāsadbhāva), with attachment to possessions (upadhi) the case is different. Its degree of infringement is unequivocal, because its character as bondage is unequivocal owing to the unequivocal presence of impure manifestation of consciousness, which is proved by its non-occurrence without that same (tasya sarvartha tadvinābhāvitvaprasiddhyad aikāntikāśuddhopayogasadbhāvasyaikāntikabandhatvena chedatvam aikāntikam eva). 102
-
In other words, whereas the taking of life is not necessarily binding because it is not necessarily connected to aśuddhopayoga, upadhi is always binding because it is necessarily accompanied by aśuddhopayoga. The Tattvadipikā goes on to say arhats and paramaḥ śramaṇāḥ (supreme śramaņas) have entirely rejected upadhi, and so it should be rejected by others too because, like 'internal negligence', it does not occur without aśuddhopayoga.103
In the Tattvadīpikā on the following gathā (3:20),
101 Trans. after Faddegon [Kundakunda (3)] p. 164 of: athaikāntikāntarangacchedatvad upadhis tadvat pratiṣedhya ity upadiśati. Faddegon renders upadhi as 'appropriation'.
102 TD on Pravac. 3:19, trans. after Faddegon.
103 ata eva caparair apy antarangacchedavat tadanāntarīyakatvāt prāg eva sarva evopadhiḥ pratiṣedhyaḥ - ibid.

Page #187

Kundakunda: The Pravacanasāra 173 Amrtacandra claims that Kundakunda teaches that the purpose of rejecting upadhi is in fact to reject internal cheda. Moreover, since internal cheda is no different from aśuddhopayoga, then upadhi is another name for aśuddhopayoga; so it is really the latter, an impure state of consciousness, which is the instrument of bondage. 104
The gāthā (3:20) commented on here reads:
If there is no renunciation (absolutely) free from (any) expectation, the monk cannot have the purification of mind; how can he effect the destruction of karma, when he is impure in mind?105
In other words, purity of mind and renunciation of worldly-objects through an attitude of indifference are one and the same thing. For, if ņiravekkho / nirapeksa is taken to be the form of cāgo / tyāga, then the emphasis falls, not upon physical renunciation, but upon an attitude of indifference, an attitude which is the same thing as purity of mind. For the monk who does not have this purity of mind, how, it is asked in the next gāthā (3:21), can he be free from mucchā / mūrcchā, ārambha / ārambha, and asamjama / asamyama (delusion, physical harm from worldly activities, and lack of control), and how, being attached to paradravya ('other substance'), i.e. the external world, the ajīva, can he ever realise (lit. gain) himself?106 And, consequently, in the Tattvadīpikā, upadhi is named as
104 tato 'suddhopayogasyāntarangachedasya pratiședham prayojanam apeksyopadher vidhiyamānah pratisedho 'ntarangachedapratişedha eva syāt - TD on Pravac. 3:20.
105 Upadhye's trans.; niravekkha / nirapekṣa = 'indifferent to worldly objects'; āsaya / āśaya = 'heart', 'mind', 'resting-place'. The construction is odd here. The sense seems to require nirūvekkhe cāge, locative absolute. Faddegon's rendering of Prākrit āsaya as Sanskrit āsrava, rather than āśaya (p.165], makes no real sense: asrava is by definition impure or, at best, neutral, whereas, in the co perfectly intelligible.
106 tadha paradavvammi rado kadham appāņam pasādhayadi - Pravac. 3:21.

Page #188

174 Harmless Souls the root cause of all these (mūrcchā, etc., and the attachment to paradravya which obstructs attainment of the śuddhātman). 107
From the above we can see that a number of technical terms have been strung, as it were, into a necklace of equivalences, with the process of internalisation as the string which holds them together. Thus upadhi is equated with parigraha, mūrcchā / moha, ārambha / himsā, and antarangaccheda (= internal hiņsā), and all these with aśuddhopayoga. Physical himsā in itself is neutral in terms of bondage, but the underlying attitude is decisive: if renunciation is incomplete, if there is attachment, then consciousness is impure, and vice-versa. In other words, parigraha, in the form of upadhi, is himsā internalised, i.e. it is aśuddhopayoga.
Because upadhi is characterised by the Tattvadīpikā as an unequivocally internal infringement, 108 an instructive comparison can be made with the use of the same term in the Tattvārtha Sūtra and the Sarvārthasiddhi . There, at Tattvārtha Sūtra 9:26, upadhi is characterised as being of two kinds, external and internal (bāhyābhyantaropadhyoḥ). The Sarvārthasiddhi comments that vyutsarga means tyāga, and that it is twofold, the giving up of external objects of attachment and the giving up of internal attachments. 109 External attachments are characterised as house, wealth, cereals, etc. (vāstu-dhana-dhānyādi), which are not appropriated (anupāttam, root dā). Internal attachments are the passions, which are natural to, or 'the dispositions of 110 the self.111
107 For the minimal exceptions to the rejection of upadhi in the technical sense of 'physical possessions' - 'possessions' which are karmically neutral because the attitude or manifestation of consciousness accompanying them is pure - see Pravac, and TD 3:223:26.
108 Introduction to Pravac. 3:19, see above. 109 bāhyopadhityāgo 'bhyantaropadhityāgaś ca - SS on TS 9:26. 110 S.A. Jain's trans., p. 266. 111 krodhādir ātmabhāvo 'bhyantaropadhiḥ - SS on TS 9:26. The

Page #189

Kundakunda: The Pravacanasāra 175 Here, the term upadhi has the external sense of being connected with possessions, similar to the meaning employed at, for instance, Pravacanasāra, 3:15. It also has the internal sense of attachment to the body and being under the control of the passions. As the Sarvārthasiddhi explains, renouncing attachment for the body has a technical sense here, of limited or unlimited periods of vyutsarga, which is not significant for the present discussion. The renunciation of passions, however, is clearly linked to Umāsvāti's kaşāya doctrine.
Comparing this with the other uses of upadhi discussed above, it is clear that, from its original meaning of physical articles of use, 112 the sense has been, to a greater or lesser degree, internalised and the meaning extended by both Umāsvāti and Kundakunda, but in different ways in accordance with their own explanations of the mechanism of bondage. Thus, in Umāsvāti's case, the internalisation is partial, and tied to the kaşāya doctrine, kasāya being instrumental in bondage. In Kundakunda (and Amrtacandra's) case, upadhi has been fully internalised - it is 'unequivocally an internal infringement' 113 - through being tied to the upayoga doctrine, in which aśuddhopayoga, a state of consciousness, is instrumental in bondage, 114
Bhāsya states that, 'external upadhi has 12 forms, internal upadhi is to do with (attachment for) the body and for passions' (abhyantaraḥ śarirasya kaşāyāṇam ca).
112 See Utt. 29:34, quoted p. 171, above. 113 TD on 3:19, intro.
114 At Pravac. 3:73, there is an obscure reference to 'external and internal(?) upadhi' (uvahim bahitthamajjhattham). Upadhye takes ajjhattha as adhyātma; Faddegon, following the TD, divides differently - madhyastha.. However, this gāthā belongs to the last five of the Pravacanasāra, which are a self-contained group and may be a further addition to the text. In any case, the TD is unambiguous in describing upadhi as 'unequivocally an internal infringement'.

Page #190

· 176 Harmless Souls
iii) Himsā and 'compassion' Kundakunda's internalisation and assimilation of himsā and its equivalents to aśuddhopayoga has consequences for the whole nexus of Jaina doctrine, consequences which are illustrated throughout this part of my work. Here, by way of concluding this section, I shall point to one specific doctrinal implication, interesting both in itself and because it illustrates the distance Kundakunda has travelled from some of the positions advocated in the earliest extant texts.
As we have seen (at 3:20, above), it is purity of mind, characterised by an attitude of indifference towards all possessions, that distinguishes the true monk. The action of such a śramaņa is free from delusion (moha) and so does not incur bondage (see Pravacanasāra 2:24-30); moreover, such action coincides exactly with the monastic rules. Mental disturbance both engenders and is engendered by himsā / parigraha, but it is the mental component, the underlying attitude, which is karmically significant, not the harming act itself. This is illustrated by Pravacanasāra 3:20c (an obviously anti-Svetāmbara gāthā, only present in Jayasena's recension), which reads:
If he accepts a piece of clothing, gourd-bowl and anything else, necessarily there is involved harm unto living beings, and there is disturbance in his mind. 115
This gāthā can be interpreted in two ways, both of which, in the context, are probably intended: first, it is implied that the taking of life causes a disturbance in the mind of the śramana, and that it is that disturbance which is to be avoided rather than the taking of life per se, which is karmically neutral (see above). Second, and conversely, the very act of violence reflects an already existing mental
115 Upadhye's trans. Vikkheva (see text) = Sk. viksepa, 'distraction', 'inattention'. See JPP pp. 38-41 on the major differences between Digambaras and Śvetāmbaras in matters of clothing, almsbowls, etc.

Page #191

Kundakunda: The Pravacanasāra 177 disturbance which is the root cause of the physical act. Either way, it is the internal attitude which is significant: the purity (calmness) or impurity (disturbance) of mind. In other words, the basic strategy for (correct) conduct, as prescribed in early Jaina doctrine, has been retained but internalised: physical inactivity, the antidote to external himsā, has been internalised to mental 'inactivity', i.e. to the attainment and maintenance of an inactive, and therefore pure, consciousness. Not even compassion must disturb this uncompromising stasis. On the contrary, since feeling clouds consciousness, compassion falls into the camp of aśuddhopayoga. One does not refrain from himsā because it is compassionate to do so, one refrains because the practice of himsā indicates internal impurity, i.e. impure upayoga which binds. Thus ahimsā achieved as the result · of a feeling of compassion is, by definition, not fully ahiņsā and cannot be fully liberating, although, in samsāric terms, it is relatively virtuous.
Kundakunda's position on compassion is made clear at Pravacanasāra 2:65, where he describes it as one of the constituents or prerequisites of śubhopayoga (which, we should remember, is ultimately a form of aśuddhopayoga):
He, who recognises the great Jinas, attends on Siddhas as well as saints and is compassionate towards living beings, has an auspicious resultant of consciousness. 116
Here the term used for 'compassion' is aņukampā.117 This occasionally occurs in the early Svetāmbara canon, for instance, at Āyāramga 2.15.4, where Indra is described as 'the compassionate god' (anukampaņteņam deveņam).118 At Ayāramga 2.2.1.8, the compassion (kaluna = karunā) of householders towards ill monks or nuns is described as
116 Upadhye's translation.
117 Compassion' / 'mercy' / 'pity' - Ardha-māgadhi koşa; cf. below, where the same term is used in the TS.
118 See Pischel, para. 397.

Page #192

178 Harmless Souls resulting in himsic activities, such as attention to their bodies, washing, the lighting of fires, etc. This idea, that compassion is essentially a lay virtue, something to be excluded from the world-view of the śramaņas, persists in later Jaina theory. However, in the earliest parts of the canon there also occurs a different attitude towards the affective in general and compassion in particular. At Ayāramga 1.5.5.4 (Suttāgame 320), for instance, we are presented with a version of the 'Golden Rule' in which the śramana is urged to practise ahimsā out of empathy (and thus compassion) for the suffering of others:
As it would be to you, so it is with him whom you intend to kill. As it would be to you, so it is with him whom you intend to tyrannise over. As it would be to you, so it is with him whom you intend to torment. In the same way (it is with him) whom you intend to punish, and to drive away. The upright man who lives up to these sentiments, does therefore neither kill nor cause others to kill (living beings) ...119
Similarly, at Āyāramga 1.4.2.6 (Suttāgame 242) the suffering of others is proved by inference from personal suffering:
First the persuasion of everyone should be ascertained, and then we will ask them severally: Is pain pleasant to you, or unpleasant? If they give the right answer, reply: For all sorts of living beings pain is unpleasant, disagreeable and greatly feared. Thus I say. 120
119 Based on Jacobi's trans., 1884, p. 50 of: tumam si nāma sa cceva jam hamtavvam ti mannasi; tumam si nāma sa cceva jam ajjāveyavvam ti mannasi; tumam si nāma sa cceva jam paritāveyavvam ti mannasi; evam (tam ceva) jam parighittavvam ti mannasi; (evam tam ceva) jam uddaveyavvam ti mannasi; amjú c' eyapadibuddhajīvi tamhā ņa hamtā ņa vighāyae - Suttāgame 320.
120 Based on Jacobi's trans., 1884, p. 39, of: puvvam nikāyasamayam patteyam patteyam pucchissāmo: ham bho pavādiyā kim bhe sāyam dukkham, udāhu asāyam? samiyā padivanne yāvi evam būyā: savvesim pāņāņam savvesim bhūyāņam savvesim

Page #193

Kundakunda: The Pravacanasāra 179
In other words, appeal is made to one's own experience, extended to others through feelings of empathy and compassion, as a reason for practising ahimsā. Here, we are probably close to the original rationale for the Jaina emphasis on ahimsa, whereas, in the later formulations, once behaviour has been formalised in a particular set of doctrinal rules, the dominating reason for ahimsā is that such practice is the only way to avoid karmic bondage; or, as in Kundakunda's reformulated doctrine, it is the reflex of an already-achieved state of internal purity which is, by definition, non-binding.
This later view is evident in the Pravacanasāra at, for instance, 1:85, which, in Upadhye's translation, reads:
False perception of things, absence of kindness towards subhuman and human beings and indulging with objects of pleasure - these are the characteristics of delusion or infatuation.
However, in the light of Pravacanasāra 2:65 (above), I prefer to follow Faddegon here and take karuṇābhāvo as meaning 'the feeling of compassion towards animals and men' (i.e. karuṇā-bhāva, rather than karuṇā-ābhava).121 This, indeed, is the interpretation that Amṛtacandra gives in the Tattvadīpikā. There, it is said that a component part of the 3 'stages' of moha is 'a feeling of compassion towards animals and men, who are simply worthy of respect'. 122 In other words, the Tattvadipika attempts to fit karuṇābhāvo into the triad, or three stages (tribhūmika), of moha (in the widest sense):123 viz. 1) false perception and karuṇā
jīvānam savvesim sattāṇam asāyam apariṇivvāṇam mahabbhayam dukkham ti bemi - Suttagame 242; cf. Ay. 1,6,5,4; 1,7,1,5; 1,7,3,2; Suy.
1,7,2.
121 Faddegon [Kundakunda (3)] p. 53.
122 tiryagmanuşyeṣu prekṣārheṣvapi kāruṇyabuddhyā - TD on
Pravac. 1:85.
123
See Pravac. 1:83-84 and TD.

Page #194

180 Harmless Souls correspond to moha (in the limited sense), 2) and 3) inclination towards / indulging with objects corresponds to rāga and dveşa.
This does seem a little strained as it stands, for why should karuņā be selected as one of two typical examples of moha? It may, therefore, be the case that in 1:85 we have a different 'definition of moha, congruent with, but unrelated to, the tribhūmika division, which may have been unknown to the author of this gāthā in its developed technical sense. There is also the possibility here that Amrtacandra, or both Amrtacandra and Kundakunda, are consciously using the fact that karunā plays such an important role in Mahāyāna Buddhism. That is to say, by including it within the circle of moha, rāga, and dveșa, they are denying the idea that ultimately there can be such a thing as 'detached compassion', compatible with wisdom and liberation; in the final analysis, karunā is aśuddha. Such an interpretation may be supported by the apparent Jaina preference for anukampā, rather than karunā, as the term for 'compassion'. In other words, it is possible that, at least when first employed by the Jains, karuņā was chosen when some special shade of meaning was required. 124
To take the contrast a little further, Rahula remarks that, according to Buddhist ethics, 'for a man to be perfect there are two qualities that he should develop equally: compassion (karuņā) on the one side, and wisdom (paññā) on the other'. 125 Compassion represents 'the qualities of the heart', such as love, charity, kindness and tolerance, while wisdom represents 'the qualities of the mind'. Thus, in the ideal Buddhist way of life, wisdom and compassion - the cognitive and the affective - 'are inseparably linked
124 It should, however, be noted that the Pāli Canon also uses aņukampā in the sense of 'compassion'. For instance, it is said that the Buddha gave his teaching 'for the good of the many, for the happiness of the many, out of compassion for the world' (bahujanahitaya bahujanasukhāya lokānukampāya, quoted by Rahula p. 46).
125 Rahula p. 46.

Page #195

Kundakunda: The Pravacanasāra 181 together', 126 for true wisdom is endowed with thoughts of selfless detachment, love and non-violence'. 127
The Jains, by way of contrast, aim for a state in which the affective can be discarded altogether; for ultimately liberation is only to be attained from a condition founded in indifference to everything not self. This is made explicit at Pravacanasāra 2:67, which reads:
Free from inauspicious manifestation of consciousness, not joined to auspicious (manifestation of consciousness) towards other substance, let me be indifferent; I meditate on the self whose essence ('self) is knowledge.
Since, as we have already noted at 2:65 (above), compassion (aņukampā) directed towards other beings is a manifestation of śubhopayoga, then it is clear that compassion can have no part in this meditation practice, the aim of which is to destroy all connection with paradravya, a connection which in itself constitutes bondage. 128
In other words, as one approaches the top of the ladder to salvation, one develops, on the one hand, an attitude of indifference, a kind of psychological stasis with regard to everything not self, and on the other, an intense concentration on, or realisation of, the inner 'knowledge' which constitutes the ātman alone. This isolation of the self from other selves, and from the world in general, mirrors, and eventually becomes, the isolation of the jiva that has attained kevala-jñāna at the apex of the universe.
Before leaving the subject of compassion, we should note that Kundakunda's categorisation of it as falling under
126 Ibid.
127 Ibid. p. 49. Rahula is, perhaps, sandpapering the joints here: such pairings are not unproblematical even for the Buddhists. For the tension between 'love' and 'self-restraint', see Gombrich 1971, pp. 320
128 esa me paradravyasamyogakāraṇavināśābhyāsah, 'This is my practice of the destruction of the causes of conjunction with other substance.' - TD on Prāvac. 2:67.
Jain. Education International

Page #196

182 Harmless Souls śubha- but aśuddhopayoga has its analogue in Umāsvāti's Tattvārtha Sūtra, where it is placed among the sātā, i.e. pleasure-causing, category of the vedaniya (feelingproducing) karmas. Tattvārtha Sūtra 6:12 reads:
Compassion towards living beings (in general) and the devout (in particular), charity, asceticism with attachment etc., concentration, equanimity, purity (= freedom from greed] - these lead to the influx of karmas that cause pleasant feeling. 129
The Sarvārthasiddhi comments: 'Compassion is the thinking in sympathy of one whose mind is moistened by favour for the suffering of others as it were one's own'.130
As has been noted above, the vedaniya-karmas are a subtype of aghātiyā karmas, whose function is to generate embodiment. Thus, although compassion may be viewed positively in terms of worldly experience and as leading to worldly happiness, ultimately it too must be abandoned in the quest for liberation, since it is responsible for the adhesion of some kind of karman and so contributes to bondage, albeit in a tenuous way (i.e. at a relatively high rung of the soteriological ladder). 131
The ritualisation of (external) ahimsā, in so far as it becomes a matter of following the rules to the letter with an attitude of non-attachment, can only further weaken the force of compassion as an agent in bringing about non
129 Trans. after S.A. Jain p. 178 of: bhūtavratyanukampādānasarāgasamyamādi yogaḥ kṣāntiḥ saucam iti sadvedyasya - TS 6:12.
130 anugrahārdrīkrtacetasaḥ parapidām ātmasthām iva kurvato 'nukampanam anukampā - SS on TS 6:12.
131 But compare this with TS 7:11 and $S, where it is said of compassion (kārunya) - i.e. 'the disposition to render assistance to the afflicted' (dinānugrahabhāvaḥ) (S.A. Jain's trans. of SS p. 195] - that it characterises the conduct of those who are able to practise non-violence and other vows to perfection' (evam bhāvayatah pūrnānyahimsādini vratāni bhavanti - SS). Cf. Paño. 144 (SBJ ed.) on active compassion. Once more, there are clearly two layers of doctrine here.

Page #197

Kundakunda: The Pravacanasāra
183
violence and liberation. Feeling with regard to other beings is at best irrelevant in these circumstances. At worst, it agitates and clouds consciousness; thus, once himsā has been internalised as the harm one does to oneself through aśuddhopayoga, i.e. through affective activity of consciousness, compassion, although undoubtedly a lay virtue, becomes just one more means of tying the ascetic to samsara. That is to say, compassion is consciousness directed towards the external world, towards paradravya. Against this there is the antidote of mental stasis or indifference, the purely cognitive, but, especially in Kundakunda, there is also consciousness directed 'inwards', at the ātman: i.e. the means becomes not just non-action (mental and physical) but insight into and realisation of the nature of the atman itself, of its essential isolation from everything else. And it is to this, Kundakunda's mechanism of liberation, that attention must now be turned. 132
132 The fact that logically liberation requires the cessation of all activity and, in the case of extreme internalisation, an undisturbed, pure consciousness, has not always been tempered in Jaina practice by what P.S. Jaini calls 'a real and active concern with the prevention and alleviation of suffering' [JPP p. 313]; nor has the spirit of anekāntavāda always informed and restrained behaviour based on such doctrines. Jaini cites the case of an 18th C. Sthānakavāsi monk [the Sthānakavāsis are themselves an offshoot of the Svetambaras], Bhikhanji, who established a sect, the Terapantha, 'based on the doctrine of total nonassistance to any living being (except mendicants)' [JPP p. 313]. The theoretical basis for this is that, by aiding or 'saving' other creatures, you become responsible for their future violence; moreover, "helpful" behavior almost always involved some interest in the result, hence brought an increase in karmic attachments' [JPP p. 314 fn. 63]. This proved unacceptable to the Jaina community at large, and ensured the isolation of the small group of Terapantha mendicants. However, as Jaini puts it, Bhikhanji exploited 'the doctrinal split inherent in any community that preaches the ideals of total renunciation and mokṣa, on the one hand, and the value of compassionate and charitable behavior (leading to heaven) on the other' [ibid.]. In some ways, the position Kundakunda advocates for those who aspire to liberation foreshadows Bhikhanji's: compassion too, whether actuated in behaviour, or as an attitude of mind, keeps one in bondage and prevents spiritual progress

Page #198

184 Harmless Souls
beyond a certain level of attainment. The reaction to the Terāpantha also highlights the readily perceived threat posed to Jaina social identity by such 'isolationism'. In Bhikhanji's case, it is the ethical norm of the lay community which is threatened; in Kundakunda's, internalisation also jeopardises ascetic practice, and thereby the entire structure of Jaina religious and social identity.

Page #199

The mechanism of liberation according to
the Pravacanasāra
6.1 Cāritra
It is a commonplace of Jaina doctrine that liberation (mokşa) cannot be achieved through samyag-jñāna - 'right knowledge' (and samyag-darśana, 'right faith') - alone: samyak-cāritra - 'right conduct', the third 'jewel' of the triad, is also required. As the very first sūtra of the Tattvārtha Sūtra puts it: samyagdarśanajñānacăritrāņi mokşamārgaḥ: 'Right faith, right knowledge, and right conduct are the path to liberation'. (Or to invert the formulation, the condition of bondage is threefold, consisting of mithyā-(wrong)-darśana, mithyā-jñāna, and mithyā-căritra.)These three together constitute a single path to liberation.2 Right conduct, moreover, is defined (by the Sarvārthasiddhi on Tattvārtha Sūtral.1) as 'the cessation of activity leading to the taking in of karmas'.3 Consequently, as Tatia explains, the attainment of perfect knowledge does not result in immediate liberation, since the latter requires perfect conduct too, and that is only attained when all activity (yoga) ceases (i.e. in the last moment before death).4
Prima facie, this would seem to act as a check on any tendency, such as Kundakunda's emphasis on jñāna, to undermine the rationale of physical, external discipline through the internalisation of practice. However, this only
See Tatia p. 151.
2 See SS on TS 1.1: märga iti caikavacananirdeśaḥ samastasya mārgabhāvajñāpanārthaḥ - 'The singular "path" is used in order to indicate that all the three together constitute the path to liberation'.
3 S.A. Jain's trans., p. 3, of karmādānanimittakriyoparamah. 4 Tatia p. 153.

Page #200

186
Harmless Souls
holds good as long as caritra itself is seen predominantly in terms of restraint from external, physical action; Kundakunda, however, makes it clear at the beginning of the Pravacanasāra that for him căritra too is essentially something internal, a matter of attitude. In this work, as will become evident, căritra is intimately linked to the attainment of jñana through dhyāna, rather than to the practice of external tapas. Thus, for Kundakunda, samyakcăritra becomes merely an augmentation to, or instrument of, samyag-jñāna, whereas the classical view is that it is samyag-jñāna which results in samyak-cāritra, and that it is the latter which is soteriologically crucial.5
Evidence of this internalisation is provided at the beginning of the Pravacanasara. There, in gāthā 1:6, it is stated that the jīva attains nirvāṇa by conduct (căritra) which has as its most important component perception and knowledge (darśana and jñāna).6 Such conduct, as the Tattvadīpikā explains, is that which is free from attachment (vītaraga). However, in the next gāthā, cāritra is given a more precise definition:
Conduct is indeed dharma; dharma is defined as equanimity [sama]; for equanimity is a modification of the self which is free from delusion and disturbance (or 'the disturbance of delusion'). [Pravacanasara 1:7]
That is to say, ideal conduct is an attitude of calmness (sama), an absence of passion, not a particular course of external, physical conduct. It is something that occurs internally, transforming the essential self (cf.1:8); i.e. caritra is internalised to a bhāva or mental state (see 1:9). Pravacanasara 1:11 goes on to say that if the self which
5 See the SS on TS 1.1: 'knowledge is mentioned before conduct, for conduct issues from knowledge' - trans. by S.A. Jain p. 4, of: cāritrāt pūrvam jñānam prayuktam, tatpūrvakatvāc căritrasya.
6
jīvassa carittādo damsaṇa
sampajjadi nivvāṇam ṇāṇappahāṇādo - Pravac. 1:6.

Page #201

: Kundakunda: The Pravacanasāra 187 has transformed itself through dharma (i.e. through its internal attitude) is united with the pure manifestation of consciousness then it attains to the bliss of nirvāṇa.?
This definition of cãritra as part of the ratnatraya may be compared with that given in the Pañcāstikāya at gāthā 115, viz.:
Right faith is belief in the way things are (i.e. in the tattvas). Right knowledge is the acquisition of correct knowledge about things (adhigama).8 Right conduct is an attitude of indifference towards things on the part of those who are on the path.
In other words, when you know how things really are you realise that they have nothing to do with the self, and so you maintain a liberating attitude of indifference towards them. 9
This lack of interest in external conduct is indicative of Kundakunda's attitude throughout these works. Moreover, it is probable that this situation obtains not so much because he takes external practice for granted but because his own development of doctrine has led him to a position where the external is irrelevant. This will become clear when we try to find a place for his understanding of the roles of jñāna and dhyāna in the standard Jaina categorization of tapas (as imperfectly schematised in the Tattvārtha Sūtra).
.
7 dhammeņa pariņadappă appā jadi suddhasampayogajudo
pāvadi ņivvāņasuham ... Pravac. 1:11. 8 On adhigama, see TS 1:3 and SS.
9 Here 'knowledge' may be being used in the vyavahāra or conventional sense, as opposed to the niscaya view of jñāna which refers to realisation of the self ('self-knowledge'); nevertheless, it is clear that knowledge and conduct are interdependent and inform each other at both levels. The crucial consideration, however, is how caritra is defined

Page #202

188 Harmless Souls 6.2 Sāmāyika
Before we deal with tapas and dhyāna directly, however, it is worth considering some of the associations which cluster around the word sama / sama, 'equanimity', and its synonyms in Jaina doctrine. This will provide some indication of how Kundakunda has arrived at his conception of ideal conduct as primarily an attitude of mind rather than specific, physical conduct.10
Throughout Jaina literature there occurs the term sāmāyika, often translated as 'attaining equanimity'.11 It is, however, possible to trace a substantial shift in emphasis and meaning between the sāmāyika of the early canon and sāmāyika as it appears in later works (in, for example, Kundakunda's Niyamasāra). An outline of this change will help us to a better understanding of the range and type of mental conduct that Kundakunda recommends in the Pravacanasāra.
P.S. Jaini writes that sāmāyika is 'first used in canonical texts with reference to a restraint (samyama) undertaken by Mahāvīra when he renounced the world'. This involved 'nothing less than the lifetime abandonment of all evil acts'. 12 This usage is evident in the following passage from the Āyāramga Sutta:
After the Venerable Ascetic Mahāvīra had plucked out his hair in five handfuls ..., he paid obeisance to all liberated spirits, and vowing to do no sinful act, he adopted the holy conduct. 13
10 See Pravac. 1:7, above. 11 See, for instance, JPP (glossary) p. 350. 12 JPP p. 221.
13 tao nam samane bhagavam Mahāvīre ... pamcamutthiyam loyam karettā siddhāņam ņamokkāram karei, karettā savvam me akaranijjam pāvakammam ti katļu sāmāyiam carittam paờivajjai. Quoted in JPP p. 17, fn. 40 - II.15.23 in Jacobi's trans.; cf. Utt. 28.32. Jacobi's trans., 1884, p. 199 is the one used here.
Jacobi's translation of sāmāyiam / sāmāyikam in this passage as holy' is criticised by Jaini as ignoring the technical meaning of

Page #203

Kundakunda: The Pravacanasāra
The important point to note here is that sāmāyika is essentially to do with physical conduct (căritra) and so with physical restraint. Moreover, this giving up of all harmful acts is, in effect, a compression of the mahāvrata, the five great vows of the ascetic: viz. to desist from himsā, anṛta, steya, abrahma, and parigraha (injury, lying, taking what is not given, unchastity, and attachment to things). This connection is pointed out by Pujyapāda in his commentary on Tattvärtha Sūtra 7:1 (which enumerates the mahāvrata) where he states that, 'From the point of view of sāmāyika, which consists of the cessation from everything blameable (i.e. harmful) the vow is one'. 14
As I have already explained, the earliest understanding of the mahāvrata was overwhelmingly in terms of physical restraint; the sāmāyika-cāritra of the Ayāramga must, therefore, be similarly external in its range and focus. However, this very wedding of sāmāyika and mahāvrata alerts us to the probability that, with the increasing internalisation of the vrata especially of ahimsā and aparigraha (they become at least as significant as attitudes or mental events as they are as physical restraints) -, sāmāyika too will be internalised. This does turn out to be the case; the difficulty lies in deciding when - i.e. in what social and religious circumstances - it occurred.
-
189
The term sāmāyika is also used for what is undoubtedly a very ancient lay practice,15 namely, the assumption of temporary ascetic status, for periods of up to one muhurta (48 minutes).16 The purpose of this ritual is, as P.S. Jaini remarks, to lead the layman 'voluntarily and irrevocably

sāmāyika-caritra and so also its wider implications [JPP p. 17, fn. 40]. 14 Trans. of: sarvasavadyanivṛttilakṣaṇasāmāyikāprekṣyā ekam
SS on TS 7:1.
vratam,
15 One mentioned derisively in the Buddhist Anguttaranikāya 1.206, quoted in JPP p. 223, fn. 42.
16
See JPP p.
223.

Page #204

190 Harmless Souls into the vows and life of an ascetic'.17 As the Āvaśyakaniryukti (c.90 C.E.?)18 remarks, during the time of sāmāyika, 'a layman becomes like an ascetic and for that reason it should be performed often'. 19
The extent to which this practice in its earliest form was fully meditational (i.e. aiming at self-realisation or gnosis through concentration), as opposed to simply a physical restraint, is not clear. P.S. Jaini does not address the problem, giving a largely synchronic analysis of sāmāyika based upon Williams' excellent study of the medieval śrāvakācāra material.20 It is reasonable to suppose that lay discipline in this respect would be meditational only to the extent that the corresponding ascetic discipline was also yogic. Yet there is also the possibility that sāmāyika, as the mental rehearsal or internalisation of ascetic practice, came to be instrumental in engendering a corresponding internalisation in the very ascetic behaviour it was attempting to concentrate. This process, if genuine, would have been facilitated by the fact that it was the same people who may have been practising in this way, as laity, who would eventually have become śramaņas. Neverthelss, it remains to be shown that the earliest practice of lay sāmāyika was anything more yogic than kāyotsarga, thought-free physical immobility.21
Sāmāyika-cāritra was not, of course, confined to lay practice. For Mahāvīra, as we have seen, it was the one all
17 Ibid. p. 226. 18 See Schubring 1962, para. 55. 19 R. Williams' paraphrase of: sāmāiyammi u kae samaņo iva sāvao havai jamhā / eeņa kāraṇenam bahuso sāmāiyam kujjā - quoted 1963, p. 133. 20 JPP pp. 221-227; cf. R. Williams 1963, pp. 131-139.
21 Eventually, sāmāyika comes to include pūjā, 'meditation by worship' and, indeed, is used as a blanket term for all types of spiritual activity. Moreover, according to the later classification of lay spiritual progress, it is the third pratimă (out of eleven), and the first of the four lay śikṣāvrata. For the relation of these vrata to each other and to mendicant practice, see JPP p. 182; see also ibid. p. 186-187, p. 190.

Page #205

Kundakunda: The Pravacanasara
191
one
embracing ascetic vow; however, the term also appears in canonical texts, not as a comprehensive vow, but simply as albeit the first - among six āvaśyaka or 'essential duties' of the mendicant, to be performed daily.22 Here, sāmāyika means something like a state of 'mental equanimity' or 'equilibrium' which leads to the end of all sinful activity.23 In other words, it is an internal state, or 'meditation', initially designed to bring about the cessation of harmful conduct. Schubring sees it as a 'state of inward balance', allied to or manifesting itself in 'an act of devotion to be repeated several times a day'.24 He characterises the āvasya as 'formulae' which have to be known, and sāmaiya as 'a short vow to be brought to one's mind repeatedly during the day promising to shun for life all that is blamable in thoughts, words and deeds ... as well as in all one has personally caused and approved of (i.e. it corresponds to the 'three restraints' or gupti).25 However, Schubring seems to be reading the term in the light of a particularly mechanised and ritualised context. For, as P.S. Jaini makes clear, these formulae are only the start of the practice which leads on to deeper meditation.26 Sāmāyika, therefore, clearly has a more general and primary sense of developing a practice of equanimity towards everything exterior to one's self, with the understanding that the corollary of such an attitude is necessarily good physical conduct. That is to say, the physical restraint connected with Mahāvira's one great vow has been partially
-
22 Indeed, before taking the full mendicant vows, the novice Śvetāmbara monk is said to live in the samãiya state or branch of the monastic order (-kappa-ṭṭhii) - see Schubring 1962, paras. 136, 138, 151. For details of the avaśyaka, see Schubring 1962, para. 151; R. Williams 1963, pp.184-5; JPP pp. 189-191.
23 See Utt. 28.32, 29 (intro.), and 29:8. Jaini says that the minimum amount of sāmāyika required of a monk is three periods a day - JPP p. 182.
24
25
Schubring 1962, p. 299; see paras. 151, 170. Ibid. p. 269. On gupti, see section 1 above. JPP p. 222ff.
26

Page #206

192 Harmless Souls
internalised, in recognition of the fact that external action is engendered and sustained by internal attitude. (The inversion of this, of course, is that if one's attitude is pure then, ipso facto, one's external behaviour will be so; the focus of personal practice has thus shifted to the former, since it is the internal attitude which is really instrumental in binding and liberating the soul.)
Pujyapāda seems to take this further in the Sarvārthasiddhi. Starting with the canonical understanding that sāmāyika is a single great vow, he equates this with the subject becoming one, or concentrated:
To become one is samaya. Sāmāyika is just the same as samaya, or it can be analysed as having samaya as its purpose.2
27
It is thus made clear that external restraint is the means to a unified internal state, a becoming 'one'. Moreover, it is this inner concentration which is the immediate cause of liberation. This is spelt out elsewhere in the Sarvärthasiddhi. At Tattvärtha Sutra 9:2, the causes of samvara for ascetics are listed; these are gupti, samiti, dharma, anuprekṣā, parīṣahajaya, and caritra (control, carefulness, virtue / duty, contemplation, victory over the afflictions, and conduct). According to Pujyapada, conduct (căritra) is mentioned last to indicate that it is the direct cause of liberation.28 And foremost, and of most importance, among these five kinds of liberation-causing conduct is sāmāyika.29
So far we have seen that sāmāyika has been characterised chiefly as the development of an attitude of
27 ekatvena ayanam gamanam samayaḥ, samaya eva sāmāyikam, samayaḥ prayojanam asyeti vā vigṛhya sāmāyikam - SS on TS 7:21.
28 cāritram ante grhyate mokṣaprāpteḥ sākṣāt kāraṇam - SS on TS
9:18.
29 See ibid.: sāmāyikādīnām ānupūrvyavacanam uttarottaragunaprakarṣakhyāpanārtham kriyate; cf. Pravac. 1:7 and Pañc. 115, quoted above, pp. 186-187.

Page #207

Kundakunda: The Pravacanasāra 193 restraint or indifference towards the world - an attitude which in turn entails physical restraint. This reading looks, as it were, over its shoulder at that bodily control which is the ultimate factor in liberation. However, the kind of concentration of and on the self, described by Pūjyapāda, looks forward to the self-realisation or jñāna of Kundakunda.30
When we turn to Kundakunda himself, we find that the only sustained use of the term sāmāyika, as such, is in the Niyamasāra (although, as we have already seen, in the Pravacanasāra there is the crucial equation of śama with cāritra and dharma). There, in the section on pratyākhyāna (renunciation), we find the following verses:
Whatever wrong conduct is in me, I give it all up together with threefold activity and practise threefold equanimity (sāmāyika) which is everything and formless, [Niyamasāra 103131
and
I am impartial (sāmya) towards all living creatures and I have no animosity towards any of them. Having given up all desires, deep meditation (samādhi) is attained.[Niyamasāra 104]32
30 It is possible, of course, that Pūjyapāda had read Kundakunda; certainly, he must have been aware of some of the latter's source material.
31 Niy. 103 trans. by Uggar Sain (Kundakunda (1)], with alterations. The surrounding verses are ślokas, but 103 is metrically defective.
32 Niy. 104. It is worth noting that commentators generally take verses 77-139 of the Niyamasāra to be Kundakunda's version of the six āvaśyaka. As Bhargava (pp. 166-167) points out, this particular list of āvaśyaka is slightly different from all other versions, before and after Kundakunda. If it is really supposed to be a list of āvaśyaka then it is a radically internalised one, in which pratikramaņa (confession), prāyaścitta (repentance), etc. are all done by the self to the self through inner discipline, and have no external indicators or emblems; that is to say, they are de-ritualised.

Page #208

194 Harmless Souls
First, it should be noted that sāmāyika is equated with inactivity: it is again connected with the threefold restraint (gupti) of mind, speech and body. Yet this is viewed not so much as a renunciatory practice as the attainment of 'everything'. It may be asked, what can be meant by a 'threefold equanimity' that is everything and formless? Uggar Sain's modern commentary33 splits the practice of sāmāyika into three tiers (lowest, middle, and highest); however, it seems clear that the threefold sāmāyika is simply the restraint of body, speech, and mind, in negation of their threefold activity, and that this practice is viewed as essentially one, a total equanimity rooted in realisation of the self's true relationship (i.e. lack of relationship) with everything else.
Second, equanimity, the condition of impartiality towards everything, is closely connected with the meditational state of samadhi, the former apparently being a prerequisite or preliminary form of the latter.
A section of the Niyamasāra, called 'parama-samādhi' (122-133), deals with this relationship in more detail. Here, in gāthā 123, parama-samadhi is seen as the result of meditation on the the self through dharma-dhyāna and śukla-dhyāna.34 When this is compared with gāthā 133, it can be seen that meditation on the self (through dharmaand śukla-dhyāna) is the practice of sāmāyika.35 Gāthā 133 reads:
In the teaching of the omniscient, he who continuously practises dharma-dhyāna and śukla-dhyāna, for him there is lasting equanimity (sāmāyika).
Taking these two gāthās together, it becomes clear that
33
Based on Padmaprabha's commentary of c. mid twelfth century C.E., on which see Upadhye p. xl, fn. 1.
34
dhammajjhāņeṇa sukkajhāņeṇa |
jo jhāyai appāṇam paramasamāhi have tassa - Niy. 123. 35 On dharma- and śukla-dhyāna, see below.

Page #209

Kundakunda: The Pravacanasāra 195 sāmāyika leads to parama-samādhi. Indeed, that they are perceived as virtually synonymous may be inferred from the fact that, barring the first two gāthās (122 and 123) and the last (133), the whole of this 'parama-samādhi' section is devoted to characterising sāmāyika. In selective summary, all external tapas and study is useless to one devoid of equanimity (124); but an all-embracing attitude of nonattachment, of sama (126), towards all things brings one close to the ātman (127) and constitutes sāmāyika. In other words, it is not so much a condition resulting from realisation of the self (of its true nature), but a means to engendering that realisation. And it is through such realisation, brought about by meditation on the true nature of the self (= sāmāyika [123]), that parama-samādhi results. Pure self-awareness or self-knowledge is inextricable from sāmāyika, and it is for this, as we shall see, that Kundakunda's ideal ascetic is striving.
It is clearly with definitions such as these in mind that P.S. Jaini refers to sāmāyika as meaning both 'attaining equanimity' and 'fusion with the true self, or as 'becoming fixed in jñāna-cetanā, pure self-awareness'.36 This, as Jaini further remarks, amounts to a 'progressive detachment of one's consciousness from all external objects';37 yet, rather than being mindless, such an attitude - as we shall see - leads, according to Kundakunda, to nothing less than omniscience.
Briefly, we have considered the shift in meaning of 'sāmāyika' - from being a synonym for the total physical restraint of the mahāvrata, via internalised restraint, to an attitude of mind, or development of consciousness through meditation, which leads to pure self-awareness. By doing so, we have charted a line of development through which the ascetic (especially, but not exclusively) has acquired a greater and greater autonomy of means vis à vis his personal liberation.
36 JPP p. 221.
37 Ibid.

Page #210

196 Harmless Souls
With this rapid sketch of one strand of doctrinal development in mind, we must now return to a more detailed analysis of Kundakunda's internalised mechanism of liberation.
6.3 Tapas and dhyāna
The ninth adhyāya of the Tattvārtha Sūtra deals with samvara and nirjarā, the obstruction of the influx of karmic material and the destruction of that already bound. Samvara is effected by gupti (restraint), samiti (carefulness), dharma, anuprekṣā (reflection), parīṣahajaya (victory over the twenty-two afflictions) and caritra (conduct of five kinds) (Tattvärtha Sūtra 9:2).38 Nirjarā is effected by tapas (austerities / penance) (Tattvārtha Sutra 9:3). The Sarvārthasiddhi adds that tapas causes both samvara and nirjarā; indeed, it is the chief cause of samvara.39 This, says Pujyapāda, is why tapas is mentioned separately here, although it is also included as a sub-category of dharma (at Tattvärtha Sutra 9:6). From this it is clear that tapas is by far the most important element in the process of achieving permanent liberation from the bondage of karman. (There are, of course, degrees of tapas, and so also of samvara and nirjarā, but not of final liberation.)
Umāsvāti divides tapas into two categories, bāhya (external) and uttara ('higher', i.e. the internal) (Tattvärtha Sūtra 9:19-20). External tapas need not detain us here, other than to remark that it consists of:
1) anaśana - fasting
2) avamaudarya - reduction in food intake
3) vṛtti-parisamkhyāna - restrictions on the begging of food 4) rasa-parityāga - rejection of stimulating or delicious food
5) vivikta-sayyāsana - sitting / sleeping in a lonely place
38 On caritra, see TS 9:18; cf. p. 185ff. above.
39
samvaram prati prādhānya pratipadanärtham - SS on TS 9:3.

Page #211

Kundakunda: The Pravacanasāra 197 6) kāya-kleśa - mortification of the body. [Tattvārtha Sūtra 9:19]
According to the Sarvārthasiddhi, these are called external 'because they are dependent on external things and these are seen by others'.40
Internal tapas consists of:
1) prayāścitta - repentance of transgressions due to negligence
(nine kinds, see Tattvārtha Sūtra 9:22) 2) vinaya - reverence, especially to elders (four kinds, see
Tattvärtha Sūtra 9:23) 3) vaiyāvsttya - respectful service to other monks, especially when
they are ill (ten kinds, see Tattvārtha Sūtra 9:24) 4) svādhyāya - study of the scriptures (five kinds, see Tattvārtha
Sutra 9:25) 5) vyutsarga - renunciation of external and internal attachments
(upadhi - see Tattvārtha Sūtra 9:26) 6) dhyāna - meditation (four kinds, see Tattvārtha Sūtra 9:28)
[Tattvārtha Sūtra 9:20]
These are called 'internal' (abhyantara) because of the restraint or limitation of the mind in these cases.41 In practice, commentators agree that dhyāna is the most significant of these internal austerities and the most important feature in the pattern of Jaina ethics.42
Dhyāna is first defined (in the Sarvārthasiddhi on Tattvārtha Sūtra 9:20) as the abandoning of mental confusion (literally, the giving up of the distractions of/ to the mind - cittaviksepatyāgo dhyānam).43 A more technical description follows (at Tattvārtha Sūtra 9:27), where meditation is said to be the concentration of thought on one point or object, lasting up to a maximum of one muhurta
40 S.A. Jain's trans. of: bāhyadravyāpekṣatvāt parapratyakșatvāc ca.
41 manoniyamamanārthatvāt - SS on TS 9:20. 42 See, for example, JPP p. 251, and Bhargava p. 193. 43 On citta-viksepa, cf. YS, 1:30, 1:31. -

Page #212

198 Harmless Souls
The
(forty-eight minutes) for the most robust.44 Sarvārthasiddhi connects such meditation with jñāna by defining dhyāna as 'knowledge which shines without quivering like the steady flame of a candle'.45
This dhyāna, according to Umāsvāti, is of four types:
1) ārta
2) raudra
3) dharmya virtuous concentration 4) sukla pure concentration.
concentration on something painful concentration on something cruel
[Tattvärtha Sutra 9:28]
The first two types (ārta and raudra) are aprašasta (not recommended) because they are the cause of the influx of inauspicious karmas.46 Conversely, the second two (dharmya and sukla) are prasasta (recommended) because they have the power to destroy karman.47 Moreover, dharmya- and śukla-dhyāna are the causes of mokṣa (Tattvärtha Sūtra 9:29).
Ārta- and raudra-dhyāna are each divided into four types [see Tattvārtha Sūtra 9:30-35], characteristic, in the case of arta, of laymen and non-vigilant ascetics (Tattvārtha Sutra 9:34), and in the case of raudra, of laymen who have and have not taken the partial vows (Tattvārtha Sūtra 9:35). An 'ascetic' who is subject to this spontaneous type of dhyāna would, ipso facto, no longer be an ascetic. 48
Similarly, the liberating kinds of meditation, dharmya and śukla, are each subdivided into four types (Tattvärtha Sūtra 9:36-44). Dharmya-dhyāna is divided into:
44 uttamasamhananasyaikāgracintānirodho dhyānam āntarmuhūrtāt. Cf. Patanjali's YS 1:2: yogaś citta-vṛtti-nirodhaḥ, 'yoga is the control of the activities of the mind-field'.
45 S.A. Jain's trans. of: jñānam eväparispandāgniśikhāvad
avabhāsamānam - SS on TS 9:27.
46
apuṇyasravakäraṇatvāt - - SS on TS 9:28. 47 karmanirdahanasāmarthyāt - ibid. 48 See SS on TS 9:35.

Page #213

Kundakunda: The Pravacanasāra 199
1) ajñā vicaya - the investigation of, or meditation on, the
teachings of the Fina - especially on what can be known only
through those teachings 2) apāya-vicaya - meditation on the loss of the true path by others,
and the means to their liberation 3) vipāka-vicaya - meditation on the effects of karman on jīvas,
and the way to liberation from karman 4) samsthāna-vicaya - meditation on the structure of the universe
and the way in which jīvas are brought to their particular position.49
The Sarvārthasiddhi explains that this kind of meditation is attainable by, or characteristic of, laymen of the fourth and fifth guṇasthānas, and ascetics of the sixth and seventh gunasthānas.50
Sukla-dhyāna is also divided into four:
1) prthaktva-savitarka-savicāra 2) ekatva-savitarka-avicāra
These two 'involve discursive concentration upon the nature of the tattvas (existents)'. Each focuses on a single existent, but, in the first, the meditator's attention 'shifts from one of the existent's countless modes to another', whereas, in the second, his attention is applied to a single mode of the existent.51 They occur between the eighth and twelfth gunasthānas, and the attainment of the twelfth guņasthāna is only possible through their negation of the passions. 52
49 TS 9:36 and SS - see also JPP pp. 252-253; and Bhargava pp. 199-200.
50 See JPP p. 253. 51 JPP p. 257, based on the SS on TS 9:39-44.
52 The terminology here is very close to that applied to the first two Buddhist jhāna; in the Buddhist case, however, reasoning / conceptual thought (vitarka) usually disappears along with discursive thought (vicāra) in the second jhāna (see Lamotte, pp. 42-43). See also 'Jhāna' in Buddhist Dictionary, and e.g. Majjhima-Nikāya 1.276. Also

Page #214

200
Harmless Souls
3) sūksmakriya-anivartin - the 'meditation' of subtle activity, in which 'all gross and subtle activities of mind and speech, as well as the gross activities of the body, are absolutely stopped'. 53
4) vyuparatakriya-anivartin - absolute non-motion, in which even the subtle physical activities - breathing, heartbeat, etc.- are stopped. These two - designated anivartin, 'that from which there is no falling back'54 - occur in the final two gunasthānas, immediately preceding physical death and final liberation.
The discussion of śukla-dhyāna, in particular, is very limited in the Tattvārtha Sūtra and Sarvārthasiddhi. A number of problems attend the relevant passages, not least the assertion, at Tattvārtha Sūtra 9:37, that the first two types of śukla-dhyāna can be attained only by those who know the pūrvas (i.e. the original Jaina canon which, at the time of the Tattvārtha Sūtra's composition, was already considered to be 'lost'). These difficulties have been considered by a number of scholars and I shall not enter into a further discussion of them here.55 Enough, however, has been presented of this standard schema of tapas and dhyāna, and its relation to the jiva's progress towards liberation, to be able to ask where, if at all, Kundakunda's soteriology of liberation through jñāna and liberation fits into this pattern. In order to answer this, we must now look at Kundakunda's gāthās on liberation in greater detail.
cf. Yoga Sūtras 1:42-44 on savitarka, nirvitarka, savicāra, and nirvicāra.
33 JPP p. 269. 54 Ibid. p. 270.
55 On śukla-dhyāna see Bronkhorst p. 179; Schubring 1962, p. 315 fn. 3, p. 316. On the contradiction between the TS and Bhäsya concerning śukla-dhyāna, see Zydenbos pp. 34-35. On the later systematisation of dhyāna, especially by Haribhadra, see Tatia pp. 283291.

Page #215

Kundakunda: The Pravacanasāra 201
6.4 Meditation in the Pravacanasāra
i) Dhyāna The first direct reference to 'meditation' (hāna / dhyāna) in the Pravacanasāra occurs at gāthā 2:59:
He, who having conquered the senses etc., meditates on the pure manifestation of consciousness (which is) the self, will not be affected by karmas. How then can the life-essentials (prāņā) follow him?.56
The prānā - the 'life-essentials' or 'animating principles' - do not, as Upadhye explains, 'form the nature of the soulstuff, but they are the indications or the signs of the presence of the soul in an embodied condition'.57 That is to say, they are the only available means by which a jīva may be detected in samsāra; or as the Tattvadipikā on Pravacanasāra 2:53 puts it, they are the reason for the vyavahāra condition of the jīvä, as opposed to its condition as it is in itself, its niscaya condition.
The prāņā are fourfold - of the senses (imdiya), of the channels of activities (bala) [viz. body, speech and mind), of duration of life (āu / āyu), and of respiration (āņappāna / ānaprāņā) (Pravacanasāra 2:54). The jiva 'lives' in samsāra (in the past, present and future) by virtue of these prāņā, which themselves originate from material substances (poggala-davva / pudgala-dravya) (2:55). As the Tattvadipikā on 2:55 is at pains to point out, the prāņā, because of their material basis, cannot reach the innate nature of the soul.58 Inevitably, given their materiality, the prāņā are seen as being both the effects and the causes of
56 Translation after Upadhye. 57 Upadhye p. 19, fn. 1 on 2:54 (trans.).
58 tan na jivasya svabhāvatvam avāpnoti pudgaladravyanirvrttatvāt - TD on Pravac. 2:55.

Page #216

202 Harmless Souls
material karman.59 The karma-tainted jīva supports prāṇā again and again (i.e. it is reborn) - prāṇā which, by their very nature as active principles, involve the jiva in further himsā,60 until it gives up the attitude of possession, of 'mine', towards external objects, especially the body.61 This, according to the Tattvadīpikā on 2:58, breaks the sequence and, having conquered the senses through this renunciation of possession, the person meditating on the pure self achieves samvara. 62 In other words, there is no rebirth, for the pure soul cannot, by definition, be embodied by material prāṇā (themselves the effect and further cause of karman); for while the conjunction with material prāṇā is the cause of the soul's vyavahāra state,63 from the niścaya view the soul is quite separate from material prāṇā. As the Tattvadīpikā puts it, the cessation of this series of material karman
accrues to him who has conquered all strange substances such as the senses, like a crystal gem withdrawn from the influence (anuvṛtti) of any support, and who abides in the perfect (kevala) and motionless self, completely pure and consisting merely of manifestation of consciousness.64
·
So rather than physical tapas, it is meditation on the true nature of the self- that it is pure consciousness - which is instrumental in bringing the cycle of material prāṇā and
59 For a full description of the mechanism of this, see Pravac. and TD 2:56-57.
60
See TD on 2:57.
61 See Pravac. 2:58.
62 See Pravac. 2:59, quoted above.
63
vyavahārajīvatvahetavaḥ - TD on Pravac. 2:59.
64 Faddegon's trans. [Kundakunda (3)] p.123, slightly altered, of TD on 2:59: sa (tu) samastendriyādi paradravyānuvijayino bhūtvā samastopāśrayānuvṛttivyāvṛttasya sphaṭikamaṇer ivātyantaviśuddham upayogamatram ātmānam suniścalam kevalam adhivasataḥ syät Note once again the way in which the jewel simile is employed: the material world colours the jiva by reflection, there is no real contact.

Page #217

Kundakunda: The Pravacanasāra 203 karma to an end. The classical idea of liberation achieved through the stoppage and shedding of material karman through tapas fades before Kundakunda's idea of release through realisation of the pure self - the ātman whose svabhāva cannot, by definition, be tainted by karman - through meditation.
This is made even clearer at Pravacanasāra 2:67, which reads:
Free from inauspicious manifestation of consciousness, not joined to auspicious (manifestation of consciousness) towards other substance, let me be indifferent (i.e. neutral); I meditate on the self whose self is knowledge.65
Introducing this gāthā, Amrtacandra characterises it as a teaching of the destruction of the causes of conjunction with other substance'. 66 By becoming indifferent to paradravya, one is released from aśuddhopayoga and becomes intent upon the self alone - a state synonymous with śuddhopayoga.67
The next gāthā (2:68) takes the form of a performative or prescriptive statement, prefaced by the nominative singular of the first person pronoun, with regard to paradravya. In other words, it gives what seems to be a paradigm of meditational practice:
I am neither body, nor mind, nor speech, nor the cause of these, (I am) neither the agent, nor the instigator, nor the approver of doers / actors.
This adds up to a radical reinterpretation of the gupti doctrine of classical Jain thought.68 Freedom from
65 Cf. pp. 181ff., above, on Pravac. 2:67, etc.
66 Faddegon's trans., p. 127, of: paradravyasamyogakāraṇavināśam; cf. TD on 2:65.
67 Paraphrase of the TD on Pravac. 2:67. 68 See above; TS 9:4; JPP p. 247.

Page #218

204 Harmless Souls bondage is no longer a matter of restraining or progressively curbing the activities (yoga) of body, mind and speech with regard to what is done, caused, or approved by oneself; now it is a matter of realising that body, mind and speech are entirely alien substances (paradravya) which, in reality, have no connection with the self whatsoever. A strict dualism applies in which the very instruments and organs of yoga (activity) are denied any connection with the essential self; therefore, what body, mind and speech do or do not do is actually irrelevant to liberation. What counts now is knowledge or realisation of the true nature of the ātman through meditation, the prerequisite of which is an attitude of indifference (madhyastha) to everything not that pure self. (The possible implications of this for Jaina orthopraxy hardly need pointing out.) Thus this gāthā (2:68) may be taken as an apophatic statement about the true nature of the self - a statement which acts, mediately and meditatively, as a means of realising that nature. The Tattvadīpikā stresses at length that body, voice and mind - characterised as acetana-dravya ('unintelligent / unconscious substance', consisting of poggala-davva / pudgala-dravya)69 - act independently of the self; they are autonomous; indeed, the body, taken in its widest sense, is an automaton. For example:
I am not the unconscious substance which is the cause of body, voice and mind; indeed, these are cause even without me as cause
and
I am not the unconscious substance which is the independent cause of body, voice and mind; indeed, these are being done even without me as agent, etc.70
69 See Pravac. 2:69-70.
70 na ca me śarīravānmanaḥkāraṇācetanadravyatvam asti, tāni khalu mām kāranam antarenāpi kāranam bhavanti ... na ca me

Page #219

Kundakunda: The Pravacanasāra 205
All this amounts to saying that the essential cause of bondage is the mistaken belief that the jīva has some connection with what is ajīva. Such a belief constitutes aśuddhopayoga, which is destroyed by the realisation of the jīva's true identity (and thus of the 'mistake') through meditation on its true nature. This seems perilously close to saying that bondage is not simply maintained by delusive behaviour - the product, among other things, of mohaniyākarman - but that it is a delusion. If the jīva, by definition, cannot act or cause action, if it cannot really have any connection with matter, how can it ever have been bound? Unsurprisingly, the Pravacanasāra does not pursue this here, but starts a technical discussion of the nature of atoms (2:70ff.). 71
ii) Dhyāna and jñāna Pravacanasāra 2:98-2:108 provides a cluster of gāthās on meditation, the nature of the self and knowledge. I have commented on the significance of some of these for Kundakunda's doctrine of liberation already;72 here I shall consider their relation to dhyāna and jñāna.
Pravacanasāra 2:98 states that to identify the self with body and wealth (i.e. with paradravya, the not-self) is to resort to the wrong road (ummaggam / unmārgam). According to the Tattvadīpikā, such identification brings about a transformation into the impure self; it is this which is the 'wrong road'.73 Thus, 'from the point of view of the
svatantraśariravānmanaḥ-kāraṇācetanadravyatvam asti, tani khalu mām kartāram antareņāpi kriyamāṇāni - TD on Pravac. 2:68.
can mean to characterise manas as 'unconscious' or the product of 'unconscious substance' is not made clear. But the physical manas of Sāmkhya, a product of unconscious praksti, and totally separate from the purusa, may provide a model here.
71 Sāmkhya, of course, has precisely the same problem, one which is inherited by Vedānta.
72 See above, pp. 140-143, 147-149. 73 aśuddhātmapariņatirūpam unmārgam - TD on 2:98.

Page #220

206 Harmless Souls impure only the impure self is attained'.74 Moreover, commenting on the previous gāthā [2:97], Amstacandra has already associated this emphasis on the impure with the vyavahāra view of the self, i.e. with the point of view that the self is and can be contaminated by paradravya.75 So for Amstacandra, at least, when it comes to the question of liberation, the vyavahāra view is not simply a theoretical construct - a partial view, or the truth at one particular level - it is actually instrumental in further bondage: to believe that the soul can be connected with and thus contaminated by paradravya is to bring about that very contamination. 76 And the self which does not cease to identify with other substances is characterised as one 'whose delusion is produced by the vyavahāra-naya'.77
'From the point of view of the pure', however, 'only the pure self is attained'.78 The person who meditates on the niścaya view - that there is no connection between the pure self and paradravya - actually becomes that pure self, i.e. he is liberated. So, according to this, it is a mental act - meditation - including or allied to another mental act - a particular kind of knowledge about the (non-)relation of the pure self to matter - which is instrumental in liberating the jiva. External means and the various kinds of material karman are disregarded. This technique is exemplified by Pravacanasāra 2:99;
He who meditates in concentration, thinking 'I am not others' and they are not mine; I am one (with) knowledge', comes to be a meditator on the (pure) self.
74 aśuddhanayād aśuddhātmalābha eva - TD, intro. to 2:98.
75 See TD on 2:97: aśuddhadyotako vyavahäranayaḥ - the conventional view which explains the impure'.
76 Although, at the theoretical level, the TD on 2:97 holds to the idea that both views are 'correct': ubhāv apy etau stah. śuddhāśuddhatvenobhayathā dravyasya pratīyamānatvāt.
77 vyavahāranayopajanitamohaḥ - TD on 2:98. 78 suddhanayād eva śuddhātmalābhaḥ - TD on Pravac. 2:99.

Page #221

Kundakunda: The Pravacanasāra 207
The Tattvadīpikā comments that the person doing this, who
lets go the non-self and, taking on the self as self, turns away from other substance and confines his thought to the single point, the self, such a one assuredly, confining his thought one-pointedly, will in that moment of confining his thought one-pointedly be pure
self.79
Here, Amrtacandra is probably writing under the influence of the synthesis of yoga traditions made by Patañjali in the Yoga Sūtras and by Vyāsa in his Yogabhāsya. Both these works predate Amộtacandra by at least several centuries. If Kundakunda himself knew the Yoga Sūtras it is not evident in his work, but since he is clearly drawing on the same ancient tradition of meditational and yogic techniques which are systematised by Patañjali, the chronology is not important. (It is interesting to note that Amộtacandra echoes not only the yogaś cittavrttinirodhaḥ of Yoga Sūtras 1:2 but with his stress on ekāgra, 'one-pointed thought' or 'concentration', recalls the technical use of this term in Vyāsa as the means by which pure samādhi is attained.) 80
On samādhi, the object of yoga, Mircea Eliade writes that, in the first place, it is 'the state in which thought grasps the object directly. Thus there is a real coincidence between knowledge of the object and the object o: knowledge'::81 According to Patañjali and his commentators, samādhi has a number of stages; by successively accomplishing these, the 'faculty of absolute knowledge' (rtambharāprajñā) is attained, and this
79 Faddegon's trans. (p.145) of: anātmānam utsrjyātmānam evātmatvenopādāya paradravyavyāvrttattvād ātmany evaikasminn agre cintăm nirūnaddhi sa khalv ekāgracintanirodhakas tasminn ekāgracintānirodhasamaye śuddhātmā syāt - TD on 2:99.
80 See Bhāsya on YS 1:1, etc. 81 Eliade p. 522.

Page #222

208 Harmless Souls
is in itself an opening toward samādhi 'without seed', pure samādhi, for absolute knowledge discovers the state of ontological plenitude in which being and knowing are no longer separated ... Fixed in samādhi, consciousness (citta) can now have direct revelation of the self (puruṣa).82
This is similar enough to Kundakunda's concentration on the pure self as knowledge - without, of course, sharing the same technical context or metaphysics - to alert us to the kind of meditational technique the Digambara writer is recommending. The general method has been discussed above (under 'sāmāyika); as a technique, it might also be compared with the preliminary form of Theravāda Buddhist meditation - samatha or samādhi, the development of onepointedness of mind.83 But such comparisons merely, demonstrate the pan-Indian character of this method; the real interest lies in the way in which Kundakunda applies it and the implications that this has for ascetic practice. And here, the fact that it bears little resemblance to (or at best subsumes) the classical Jaina modes of dhyāna as outlined in the Tattvārtha Sūtra (see above), with their emphasis on the cessation of all activity, is significant. For Kundakunda has developed a path to liberation which, at least in its later stages, is almost totally hermetic or self-referential. The goal is achieved through the individual's inner concentration on his pure self; this brings about knowledge or realisation of that self which, since the pure self has absolutely no connection with other substances (karman, etc.), is synonymous with liberation. The old, semimaterialist model of a soul which is weighted down by material karman shedding that karman through physical austerities and, in its liberation, ascending to the topmost part of the universe, has been (at least temporarily) superseded.
This becomes clear if we consider the ways in which
82 Ibid. 83 See, for instance, Rahula p.68.

Page #223

Kundakunda: The Pravacanasāra 209 meditation is characterised in the Pravacanasāra. Gāthā 2:104 reads:
He, who has destroyed the impurity of delusion, who has no interest in the objects of the senses, and who, having restrained his mind, is fixed in his own nature, is a meditator on the self.
In other words, meditation on the (pure) self is nothing less than realisation of that self; the successful meditator becomes the pure self which is his own nature (svabhāva). As the Tattvadipikā on 2:104 puts it: 'Thus meditation, which takes the form of absorption in one's own nature, is the self, because it is nothing other than the self.84
That is to say, the ātman has and can have no dravya for its substratum (adhikarana) other than its own nature; its own nature (as defined by the Tattvadīpikā) is 'infinite, innate intelligence' (anantasahajacaitanya), and it is the fixing of oneself in this svabhāva which constitutes meditation.85
Kundakunda himself has already defined the self's svabhāva in some detail as being constituted of jñāna and darśana, an object beyond the senses, eternal (dhuva), unmoving (acala), without support - so independent (aņālamba), and pure (suddha).86 This is the eternal self, whose 'self is upayoga (2:101), meditation upon which destroys moha,87 and leads to 'imperishable happiness'.88
Kundakunda then asks the question (Pravacanasāra 2:105), what does the person who has realised his pure self, who has attained kevalajñāna, meditate upon (given that he
84 ataḥ svabhāvāvasthānarūpatvena dhyānam ātmano 'nanyatvāt dhyānam ātmaiveti - TD on 2:104.
85 See TD on 2:104. 86 At Pravac. 2:100-101. 87 See above, pp. 140-143, 147-149.
88 sokkham akkhayam / saukhyam akşayam - 2:103. Note that the TD on 2:102 defines dhyāna as ekāgrasamcetana, 'one-pointed awareness'.

Page #224

210 Harmless Souls is omniscient and, as the Tattvadipikā puts it, does not feel any desire, curiosity or doubt)?89 The answer is that he meditates on 'supreme happiness' (param sokkham 2:106) which, according to the Tattvadīpikā (on 2:106), is as much as to say that the self 'continues as simply a one-pointed awareness in a state of calm'.90 This is the attainment of 'perfection, whose svabhāva is innate knowledge and bliss'.91
In other words, meditation is not only the instrument of liberation but it also characterises the state of the liberated: path and goal constitute a single practice.
iii) Jñāna It has been shown that knowledge about the true nature of the self, combined with meditation on that nature, constitutes Kundakunda's path to liberation. However, given the omniscience of the arhat, knowledge occupies an even more central place in Kundakunda's soteriology than the above might at first suggest. For he equates knowledge with the knower (i.e. the self); they are co-extensive and omnipresent. This formula, and the relation of knowledge to the objects of knowledge, must now be considered in greater detail.
Śruta Skandha 2 of the Pravacanasāra (jñeyatattvaadhikāra) ends with the following gāthā:
Therefore, having thus realised that the self is innately disposed to be a knower, stationed in unpossessiveness, I turn away from the idea of 'mine'. [Pravacanasāra 2:108]
The Tattvadīpikā comments:
89 abhilașitam jijñāsitam samdigdham - TD on Pravac. 2:105. 90 anākulatvasamgataikāgrasamcetanamātrenāvatisthate - op. cit.
91 sahajajñānānandasvabhāvasya siddhatvasya - ibid. For the equation of 'happiness' with knowledge, omniscience and liberation, see Pravac. 1:59 and Upadhye's footnote on p. 8 of his translation of 1:59.

Page #225

Kundakunda: The Pravacanasāra 211
This I, qualified for liberation by means of adhering to the idea of 'not mine' and abandoning the idea of 'mine', preceded by full aquaintance with the truth that the self is inherently a knower, devotes itself to the pure self in all its undertakings because there is no gap between them.
That is to say, I am indeed inherently a knower, and as I am one whose knowledge is absolute (i.e. as I am omniscient) I have a relationship - which takes the form of that between a knower and the naturally knowable - with everything, and no other relationship, such as that between possessor and possession. I am, therefore, unpossessive towards all things and attached to nothing. 92
What does it mean to say that the ātman is naturally or inherently a knower (or as the Samayasāra puts it of the emotions and the operation of karman: 'they are not my own nature; I am exclusively [uniquely) a knower by nature')?93
In the first instance, this is a perception which relies upon the teaching that the liberated soul does not simply attain a condition of isolated bliss at the apex of the universe, it is also characterised by omniscience (kevalajñāna). (Although, for the sake of precision, one should distinguish between those liberated souls who have already discarded their bodies (siddhas) and those arhats who have attained omniscience but for the time being remain embodied.) P.S. Jaini renders the term 'kevalajñāna' as 'knowledge isolated from karmic
92 aham esa mokṣādhikārī jñāyakasvabhāvātmatattvaparijñānapurassaramamatvanirmamatvahānopādānavidhänena krtvāntarasyābhāvāt sarvārambhena śuddhātmani pravartate tathāhi - aham hi tavat jñāyaka eva svabhāvena, kevalajñāyakasya ca sato mama viśvenāpi sahajñeyajñāyakalaksaņa eva sambandhaḥ na punar anye svasvāmilakṣaṇādayaḥ sambandhāḥ|| tato mama na kvacanăpi mamatvam sarvatra nirmamatvam eva | TD on Pravac. 2:108.
93 na du te majjha sahāvā jānagabhāvo du aham ikko - Sam. 198 [= 213); cf. ibid. 207.

Page #226

212 Harmless Souls
interference'. He goes on to say that such knowledge is
compared to a mirror in which every one of the innumerable existents (dravya), in all its qualities (gunas) and modes (paryayas), is simultaneously reflected. These 'knowables' are cognised without any volition whatsoever on the part of the arhat. Furthermore, no activity of senses or mind is involved; there is only direct perception by the soul.94
Omniscience is thus the natural state of the soul, in the sense that, when all karmic obstruction is removed, that kind of knowledge automatically obtains. To put it another way, omniscience is not something to be striven for or attained, in the sense of being some quality which is added to or gained by the agent or knower, rather it is something to be realised or revealed (through the shedding of karman) as the original nature of the self. Moreover, since the jīva has, in reality, no physical relation with anything ajīva - their relation is that of the knower and the knowable -, it is apparent that in its fundamental nature the soul has never been anything but omniscient. And to achieve that omniscience it is only necessary to realise it, through meditation on the true nature of the self and its relationship with the world of matter. Again this seems tantamount to saying that karmic bondage - the adherence of matter to the jiva is fundamentally unreal, a mistake or delusion. For while on one level the realisation of the original nature of the self shines through when obstructive karman is removed, on another it is that very realisation which is instrumental in removing obstructive karman, through the perception that in reality karman cannot obstruct the pure self.
Referring to Kundakunda's Niyamasära (159 = 158 SBJ ed.), P.S. Jaini remarks that the defining mark of the omniscient being is 'complete self-knowledge ...; any other
94
JPP p. 266.

Page #227

Kundakunda: The Pravacanasāra 213 description is simply a worldly or "conventional" one'. 95 That is to say, 'From the vyavahāra point of view the omniscient lord sees and knows everything, from the niścaya point of view the omniscient sees and knows the self.96 But, according to the Pravacanasāra, this exclusive knowledge of the self is not a limitation on or contraction of knowledge; on the contrary, knowledge of the self includes knowledge of everything else. As gāthā 1:23 puts
it:
The soul is co-extensive with knowledge; knowledge is said to be co-extensive with the objects of knowledge; the object of knowledge comprises the physical and non-physical universe; therefore knowledge is omnipresent. 97
It follows from this that the soul too is omnipresent. Although this is admitted as a temporary possibility in certain circumstances by orthodox doctrine, it is not seen as a characteristic of the liberated soul.98 Kundakunda, however, confirms at Pravacanasāra 1:26 that this is how he understands the nature of the kevalin:
The great Jina is everywhere and all the objects in the world are within him; the Jina consists of knowledge, and those referents of words (i.e. objects) are declared his because they are the objects of knowledge.
As the Tattvadīpikā on 1:26 states, this means simply that the Jina knows all the objects in the world completely (i.e. he knows the meaning of all words and so he knows
95 Ibid. p. 267.
96 Niyamasāra 159. Cf. Niyamasāra 166: appasarūvám pecchadi loyāloyam ņa kevali bhagavam ... - '[From the niscaya point of view] the omniscient lord sees the real nature of the self, not the universe and non-universe....
97 Upadhye's translation, p. 4. 98 See JPP pp. 102-3, and especially p. 269.

Page #228

214 Harmless Souls their referents). In reality, he has no physical or metaphysical contact with them, or they with him.99 That is to say, there is no activity on the part of the liberated self, or - as the Tattvadipikā points out - it never leaves its svatattva. 100
This inactivity of the self includes 'knowing'; indeed, there is no 'knowing' as such (in the sense of process) for the pure self; for, since the knower (i.e. the self) has knowledge as its own nature, 101 it does not have to do anything in order to know apart from realise its true nature. This relationship is confirmed by gāthā 1:27:
The doctrine is proclaimed that knowledge is the self; without the self there is no knowledge. Therefore, knowledge is the self; but the self is knowledge or anything else. 102
(The Tattvadīpikā explains that the self, 'being the seat of innumerable properties', 103 may be knowledge owing to the dharma of knowledge, or owing to some other dharma it may be something else. 104 In other words, the self and knowledge are co-extensive, but the latter does not define the former, which is actually and potentially much greater.)
The non-knowingness of this ātman (= jñāna) is further
99 Sarve 'rthās tadgatā ity upacaryante, na ca teşām paramārthato 'nyonyagamanam asti, sarvadravyāņām svarūpanişthatvāt - .... objects are said metaphorically to belong to him ... but in the real sense of the word there is no mutual going towards each other, since all substances abide in their own characteristic-nature' - Faddegon's trans., p. 16.
100 Ibid. Indeed, at one level, ontological description as such is probably irrelevant to Kundakunda's purpose here which, like that of much, if not all, 'mystical' teaching, is to engender in the audience a particular attitude or transformation of consciousness. In other words, the teaching is itself directly instrumental in self-realisation: conventional or partial knowledge helps to liberate absolute knowledge.
101 ņāņi nānasahāvo / jñānī jñānasvabhāvaḥ - Pravac. 1:28. 102 Translation after Upadhye p. 4.
103 Faddegon's trans., p. 16, of anantadharmādhisthānatvāt - TD on 1:27.
104 jñānam anyadharmadvāreņānyad api syāt - ibid.

Page #229

Kundakunda: The Pravacanasāra 215
stressed by Kundakunda:
He who knows is knowledge; the self does not become a knower through knowledge; knowledge develops of itself, and all objects are found in knowledge. 105
As Faddegon puts it in his translation of this gāthā, 'the self does not by the help of its knowledge become somethingthat-is-knowing (jñāyaka)';106 knowledge is thus the natural or revealed state of the karmically unobstructed ātman. The Tattvadipikā explains that this equation obtains because 'the self is an actor of the greatest supremacy and power in whom agency and instrumentality are united'. 107 In other words, the pure self does not do anything; despite the confusion of terminology it is not an 'actor' in the ordinary sense of that word, it simply is.108 Moreover, all the objects of knowledge, since they are said to be found or 'stand' (tthiya/sthita) in knowledge, are thus also found in the self which has been equated with knowledge. 109
It is in this context that passages such as those quoted from the Niyamasāra (159; 166 see above) should be understood. The pure self or kevalin knows everything without coming into possessive relation with anything notself.110 Knowledge is essentially a matter of indifference or non-attachment, not of possession. That is to say, it is not the result of a process ('knowing') but a permanent state which is revealed and realised through meditation on the true nature of the self. Such meditation focuses on the status of the relationship between jiva and ajīva, the knower and the known; the full realisation of that
105 Translation after Upadhye p. 5.
106 Faddegon's trans., p. 21.
apṛthagbhutakartṛkaraṇatvaśaktipāramaiśvaryayogitvād
107 ätmanaḥ - TD on 1:35.
·

108 Cf. Niyamasāra 172.
109
Pravac. 1:23, 1:26-7, 1:36: tamhā ṇāṇam jīvo.
110
See 2:108 and TD, above.

Page #230

216 Harmless Souls relationship is liberation and omniscience. For Kundakunda, an understanding of this is clearly the crucial component in his mechanism of liberation. He reiterates the relationship in a number of similar formulations. For instance,
The omniscient lord neither grasps nor releases nor transforms the other; he sees all around and knows everything completely, 111
and
The knower, who is beyond sense-perception, necessarily knows and sees the whole world neither entering into nor entered into by objects of knowledge, just as the eye sees the objects of sight. 112
The Samayasāra uses a different image to illustrate the relation of the knower (i.e. the self) and the object known (from the niscaya view), but with the same meaning:
Just as chalk does not become the other (i.e. the surface it is applied to) but remains chalk qua chalk, so the knower does not become the other i.e. the object known) but remains the knower qua knower. [Samayasāra 356 (= 385 J.L. Jaini's ed.)]113
And again, in a familiar image of the contactless, noncontaminating relation of self and not-self, the Pravacanasāra states that 'knowledge operates on objects just as a sapphire, resting on milk, pervades the whole of it with its lustre'. 114
111 Translation after Upadhye, p. 4, of Pravac. 1:32.
112 Upadhye's trans., p. 4, of Pravac. 1:29. Upadhye takes na pavittho nāvittho as Sk. na pravistaḥ na āvistaḥ, as against the commentators who take na avistaḥ (na apraviştah) - see fn. 2, p. 4.
113 Cf. Pravac. 1:28 and 1:29, above, where the objects of knowledge are compared to the objects of sight: they are within range of the knower / seer, but, crucially, there is no 'mutual inherence' (Upadhye's trans., p. 4, of nevaņnonnesu).
114 Upadhye's trans., p. 4, slightly altered, of Pravac. 1:30.

Page #231

Kundakunda: The Pravacanasāra
217
To summarise, one becomes such a knower by rehearsing in meditation this attitude of detachment - which springs from knowledge of the true relation between self and other towards everything not-self (ajiva / para). The understanding that the self is a knower rather than a possessor leads to the abandonment of a possessive relationship with anything not-self - indeed, from the niścaya point of view, such a relationship is an impossibility, i.e. a delusion, anyway. For, in reality, there is nothing the pure self can do except be itself. In other words, kevalajñāna points to the isolation and inactivity of the self; the atman is co-extensive with and yet not of the world; it is rather than does. To such an entity, karman and the fruit of karman are, in the final analysis, irrelevant; they do not bind what, in reality, cannot be bound. Selfknowledge is the sole key to liberation.
6.5 The rationale for external, ascetic practice, according to the Pravacanasāra
In the light of this stress on self-knowledge, or selfrealisation, and meditation, the question arises of what rationale can be offered for continuing with external practices, i.e. with the identity-defining practices of the Jaina ascetic. Perhaps because his works were composed primarily for those who were already habitually ascetic in their practice, Kundakunda does not address this problem directly. However, in the Pravacanas āra and its commentaries there are a small number of significant references which indicate a recognition albeit a philosophically unsatisfying one - that some kind of answer is required.
In gāthās 3:5-6, Kundakunda lists some characteristics or 'emblems' (linga) of the Jaina ascetic. First, he states that:
The mark (of a Jaina ascetic) consists in possessing the form in which one is born, in pulling out hair and beard, in being pure, in

Page #232

218 Harmless Souls
not harming beings, etc., and in not attending to the body. 115
Then he adds:
The Jaina mark, which is the cause of the stoppage of rebirths, consists in being free from action based on delusion, in being endowed with purity of manifestation of consciousness and purity of activity, and in being independent of the other (everything not
self).116
Considering 3:5 first, the expression 'possessing the form in which one is born' (jadhajādarūvajādam) occurs in a slightly different formulation in the previous gāthā (3:4), as jadhajādarūvadharo, 'wearing a form similar to that in which he is born'.117 Upadhye explains that this means that the person wishing to be an ascetic (the subject of these gāthās) 'should give up everything including clothes and remain naked; this is the excellent type of Jaina asceticism'.118 In other words, this is equivalent to the English colloquial expression 'wearing one's birthday suit', meaning going completely naked (which, of course, is the most obvious characteristic of the Digambara ascetic).
The other components of 3:5 are self-explanatory, except for śuddha. The Tattvadīpikā describes this 'purity' in material terms, as being due to the negation of 'possessing anything'.119 As we have seen, in the Sarvārthasiddhi (on Tattvārtha Sūtra 9:6) 'purity' (sauca) is glossed as 'freedom from greed' (parigraha I lobha), the defining characteristic of the householder's way of life; 120 it initiates himsā and causes bondage. In short, purity both
115 Trans. after Upadhye p. 25; see ibid. p. 25 fn. 2, on pratikarma, which he takes as a-pari-karma, a reading followed in this translation.
116 Translation after Upadhye p. 25. 117 Upadhye's translation. 118 Upadhye p. 25 fn. 1. 119 Translation by Faddegon p. 157, of sakimcanatva. 120 See above, p. 76ff..

Page #233

Kundakunda: The Pravacanasāra 219 results from and exemplifies non-possession (aparigraha ).
Suddha appears again in the next gāthā (3:6), but this time it is connected with upayoga ('manifestation of consciousness') and yoga ('activity'). In the Tattvadipikā these two are linked as cause and effect: the possession of aśuddha-upayoga, doloured by śubha- and aśubhaupayoga, engenders aśuddha-yoga (of body speech and mind); conversely, the negation of aśuddha-upayoga leads to purity of upayoga and so of yoga. In other 'words, it is the internal condition which informs external behaviour..
The Tattvadipikā claims that these two gāthās (3:5 and 3:6) describe a pair of linga, 'external' and 'internal', which together characterise the ascetic. 121 There are, however, a number of indications that, rather than a systematic definition of what constitutes śramaņa-status, we have here a conflation of two different classificatory systems, representing two different historical moments in the development of Jaina doctrine. That is to say, there is a certain amount of overlap between the content of the two gāthās which points to a separate origin for each. For instance, although in general the two uses of 'purity' discussed above divide into 'external' and 'internal', we may note that non-possession of material goods springs in the first place from an attitude - freedom from greed, and that, on the other side, two of the aspects of yoga, activity of body and activity of speech, are material in nature and external in operation. Moreover, mucchă-ārambhavimukkam / mūrcchā-ārambha-vimuktam (3:6), which I have translated as 'freedom from action based on delusion' (i.e. from action based on delusion about the real nature of the self, and thus about its relation to the non-self), is virtually equivalent to the 'purity' (suddha) listed as an 'external' aspect of the ascetic at 3:5. For as I have noted above, 122 ārambha is closely associated with parigraha in Jaina thought, and denotes a violent action initiated by
121 bahirangāntarangalingadvaitam - TD intro. to Pravac. 3:5. 122 See pp. 5, 31, etc., above.

Page #234

220 Harmless Souls greed and possessiveness. Amrtacandra's reading of mürcchā-ārambha-vimuktam, as negation of the development of the karmic consequences of the attitude of possession123 emphasises only one half of the classical theory of the equivalence of parigraha and mūrcchā as defined at Tattvārtha Sūtra 7:17 and Sarvārthasiddhi,124 where mürcchā is both the activity of preserving or acquiring possessions and an attitude of possessiveness towards them.
It is clear from these instances that neither division of the characteristics of the ascetic given at Pravacanasāra 3:5 and 3:6 falls exclusively into the 'external' and 'internal' categories which the Tattvadipikā attempts to impose. 125 I suggest that, historically, 3:5 represents an earlier understanding of the linga of an ascetic, while, given that upayoga appears to have originated with him, 3:6 is Kundakunda's revision or further internalisation of that earlier definition. It should also be noted that the next gāthā (3:7) begins with the words 'having taken this characteristic' (ādāya tampi limgam), singular, apparently referring (pace Amstacandra) to only one set of characteristics. Again, this suggests that 3:6 represents an interpolation by Kundakunda into a traditional description of the way in which one becomes a śramaņa. Moreover, Kundakunda leaves us in little doubt as to which set of characteristics he considers the more important, since he says of the highly if not totally internalised group which make up the Jaina-linga at 3:6 that they are 'the cause of the stoppage of rebirths' (apunabbhavakāraṇam).
Kundakunda's view on the relation of internal purity and external practice is, as we have seen, that the physical practice is only of value insofar as it is informed by and proceeds out of the correct internal attitude. To put it
123 mamatvakarmaprakramapariņāma - TD on Pravac 3:6. 124 See above, pp. 73ff.
125 This, of course, is not unexpected, given that Book 3 of the Pravacanasāra, in particular, appears to be of a compilatory nature.

Page #235

Kundakunda: The Pravacanasāra 221 crudely, although without significant distortion, if one takes care of the upayoga the yoga will take care of itself. This returns us to the question of why it should be thought necessary to continue with external practice at all, since the latter is only the reflection of an already-achieved internal state, and it is the internal state which is significant for one's personal liberation. Yet this very phenomenon - that the external reflects the internal - points towards an answer.
Pravacanasāra 3:5-6 deals with the linga, the characteristic mark, of the ascetic. In logic, a linga is 'the invariable mark which proves the existence of anything in an object';126 i.e. given this characteristic or characteristics, there can be no doubt about the nature of the thing examined. In this case we have to do with the linga which prove that somebody is a true ascetic and so on the route to liberation. The constituents of the external linga - nakedness, the pulling out of hair and beard, nonpossession, ahimsā, neglect of the body - are, once defined, clear enough and physically evident to anyone who knows what they are looking for. The constituents of the internal linga, however, are - from the very fact that they are internal, relying as they do on pure upayoga and attitude - not evident to observers; i.e. although they may provide an ideal towards which the individual strives, they are insusceptible to outside verification in themselves. And if they are not observable characteristics then, it may be asked, in what sense can they be linga?
The internal are, however, not totally unobservable, for, since external practice is said to reflect an already-achieved internal state, the external linga of ascetic behaviour imply the inevitable although invisible presence of the pre-defined internal linga. In other words, there is really only one linga (or set of linga) in the strict sense of the term, the external. From the presence or absence of that mark any observer can infer the presence or absence of internal states. And since it has been laid down that what really counts in terms
126 Monier-Williams, Sanskrit Dictionary.

Page #236

222
Harmless Souls
of liberation is precisely the internal state, then it becomes clear that, in this teaching, the function of external practice is to provide a linga - i.e. evidence or proof of the soteriologically crucial internal condition. 127 That is to say, it is a public demonstration or assurance, chiefly to others but perhaps also to oneself, that one is on the correct road - a visible emblem and reinforcement of Jaina identity.
Nevertheless, from the advanced ascetic's point of view, Kundakunda's stress on the internal makes external conduct This is karmically (i.e. soteriologically) irrelevant. graphically illustrated by Pravacanasara 3:27, which reads:
He whose self is non-desiring, that is asceticism that is what ascetics seek. Other food, obtained as alms, is not desired (not food); so those ascetics are not takers of food. 128
M
1
It seems that there is a pun here on eṣaṇa which can mean either 'desire' or 'alms begged in the correct manner'. The self without desire is foodless, therefore 'other' (i.e. material, actual) food is not food, in the sense that it is not desired (by the desireless self who is the ideal ascetic). In other words, for the true ascetic, external conduct - physical tapas becomes irrelevant: if he is internally pure then what he does physically can have no karmic effects for him. The Tattvadipikā bears this out:
Since in essence he is abstinent from food (anasana) and since the alms are devoid of the [fault of desire (eṣaṇa- doșa)] fault against the eṣaṇa-samiti, the self-controlled in food is visibly actually foodless. Thus: - if a man is at all times conscious only of the self, which is exempt from the taking of material sustenance, his
any

127 See above, pp. 160-163, for further comments on the external being an emblem of the internal, ref. TD on Pravac. 3:17.
128 Pravac. 3:27. Cf. Samayasara 405ff. Cf. also Pāli anesanā: 'improper alms begging' ('a wrong going for food' - CPD); e.g. DN III 224.25 - anesanam appaṭirupam āpajjati. The pun that is reflected in my translation is explained in the following paragraph.

.

Page #237

Kundakunda: The Pravacanasāra 223 innate nature is in itself foodless, inasmuch as it is void of all longing for food. This, namely, is his abstinence from food, his asceticism; for the internal is of more importance (than the external). 129
Anaśana is given at Tattvārtha Sūtra 9:19 as the first of the 'external austerities' (bāhyam tapaḥ).130 Clearly, therefore, Kundakunda and Amộtacandra are here emptying external tapas of any independent soteriological value; for if the eşaņā-samiti is not infringed, the attitude of nondesire makes even feeding not count as feeding.
This is extreme, but, given the way in which Kundakunda ascribes new values to the vocabulary of begging and fasting, we may surmise that he is here consciously reacting against excessive formalisation, the mechanistic pursuit of physical austerity. That is to say, by giving the 'true' definitions of foodlessness and alms, he provides a corrective to the mindless repetition of a practice whose underlying significance has been forgotten. From the standpoint of personal liberation this may be very necessary, but if taken as a general, social principle it clearly undermines, probably fatally, the rationale for external, ascetic practice. From the point of view of the Jaina community as a whole, it is therefore essential that some reason should be given for continuing with external tapas. Such a reason has been outlined above; it may seem philosophically weak but it is socially indispensable. The importance of public display in any religion should not be underestimated; indeed, it is what most obviously characterises a 'religion' as opposed to a private soteriology. Thus, in abstract terms, there is probably a pendulum-like
129 Faddegon's trans., p. 169, slightly emended, of: svayam anaśanasvabhāvatvād eşanādoşaśūnyabhaiksyatvāc ca yuktāhāraḥ sākṣād anāhāra eva syāt / tathāhi - yasya sakalakālam eva sakalapudgalaharaṇaśūnyam ātmānam avabudhyamānasya sakalāśanavrsnāšūnyatvāt svayam anaśana eva svabhāva / tad eva tasyānaśanam nāma tapo 'ntarangasya balīyastvāt iti ...
130 See above, pp. 196ff., and JPP pp. 250-251, on these.

Page #238

224 Harmless Souls movement or continuous balancing act in Jainism between excessive formalisation, with its threat to the personal or soteriological, and excessive internalisation, with its threat to the social or corporate identity. Given the compilatory nature of early Jaina texts (including the Pravacanasāra), the two extremes may frequently be reached in the same text. It fell to the scholastic commentators to attempt to compensate for and correct this polarisation. The doctrine of the two naya, if not conclusive, is at least a holding operation in this struggle. In terms of the present analysis, the vyavahāra-naya would thus embody the social view while the niscaya-naya would represent the personal or soteriological perspective.
6.6 Socio-religious roles in the Pravacanasāra
It remains to comment briefly on the Jaina socio-religious hierarchy implicit in the Pravacanasāra, and on Kundakunda's attitude to the laity. (The purpose of this section is simply to make explicit what may be readily inferred from the material already treated.)
The individual's place in the religious hierarchy, and thus his social role, is decided for Kundakunda by his state of consciousness. It is always a particular upayoga (manifestation of consciousness) which underlies and informs any external 'emblem' of religious status; and, in theory, it is to the underlying internal state that any question about an individual's status should be referred. In practice, the problem of gauging 'inner-states' leads, of course, to reliance upon external indicators. Nevertheless, Kundakunda is unequivocal about the meaninglessness of external practice unless it derives from and is informed by internal purity. 131
The equations are as follows: śuddha-upayoga is the
131 See above, passim.

Page #239

Kundakunda: The Pravacanasāra 225 internal state of the ideal śramana, and leads to liberation;132 aśuddha-upayoga covers the full range of inner states short of śuddha-upayoga, and leads to a relatively better or worse rebirth. Thus, in terms of socioreligious roles, the latter state characterises virtually the whole community. As we have seen, however, aśuddhaupayoga is divided into śubha- and aśubha-upayoga. śubha is clearly better than aśubha, but how positively or negatively the former is viewed depends entirely upon what audience is being addressed. Thus, in Book 1 of the Pravacanasāra, which is aimed predominantly at śramaņas, anything less than the attainment of suddhaupayoga is considered unsatisfactory; in Book 3, however, which has a more general (perhaps one should say, less ideal and more realistic) audience in mind, the cultivation and attainment of śubha-upayoga can, in itself, be a laudable activity and goal.
Subha-upayoga is clearly the inner state most open to interpretation and ambiguity in terms of the spiritual value and thus the religious status attached to it. In Pravacanasāra Book 1, śubha's aśuddha nature is stressed (1:69-1:79): it entails desire, contingency and impermanence, hankering after sense-pleasures, attachment to objects which are dependent and impermanent, and thus it results in misery (see 1:74-77). Even the gods are caught in this trap of contingency (1:71-74). And 'if men, hellbeings, subhuman beings, and gods suffer misery, born from the body, then of what use is (the distinction of) auspicious or inauspicious manifestation of consciousness for souls?'133 That is to say, if śubha- and aśubha-upayoga lead alike to rebirth and misery, what is the point of
132 See Pravac. 1:14; cf. 1:11, 2:103. 133 Translation of Pravac. 1:72.

Page #240

226 Harmless Souls distinguishing between them?134 The point is clearly to devalue śubha-upayoga in the (potential[?]) śramana's mind, and to stress that liberation can only be achieved through śuddha-upayoga. This is given full emphasis in gāthās 1:77 and 1:78.
Gāthā 1:77 reads:
He, who does not think that there is no difference between merit and demerit, wanders about in terrible, unbounded samsāra, covered in delusion.
In other words, punya and pāpa, associated with śubha- and aśubha-upayoga, are both characterised as totally samsāric. (Note the emphasis again on knowledge and delusion as the liberating and binding factors.)
In contrast to this, the condition of the aśuddhaupayogin (= the ideal śramaņa) is described in the next gāthā [1:78]:
He who, understanding the nature of things, does not experience attachment or aversion towards objects, his manifestation of consciousness being pure, destroys the suffering which arises from embodiment.
To put this in the terms of the present argument, here Kundakunda is criticising the inner-condition of the laity, probably for a śramaņic or potentially śramaņic audience. That it is indeed the laity he associates with aśuddhaupayoga is evident from Pravacanasāra 1:69, where the ātman which is characterised by śubha-upayoga is described in terms of activities associated with lay vrata:
The self which is attached to the worship of gods, ascetics and teachers, to giving, to good morals, and to fasting, etc., is a self
134 See TD on Pravac. 1:72.

Page #241

Kundakunda: The Pravacanasāra with auspicious manifestation of consciousness.135
227
In this context (Book 1 of the Pravacanasāra), therefore, lay-practice is viewed in a negative light.
However, in Book 3 of the Pravacanasāra, Kundakunda is apparently much more accommodating to śubha-upayoga, and thus, by implication, to the laity (although again he addresses himself directly not so much to the latter as to the average śramana). For here it is recognised, realistically, that śubha-upayoga, and thus a better rebirth, is also a legitimate goal - albeit a lower one - for ascetics.
Gāthā 3:45 reads:
In (our) religion there are ascetics who have pure manifestation of consciousness and those who have auspicious manifestation of consciousness. Among them, those with pure manifestation of consciousness are not subject to the influx of karmic matter, while the rest are subject to influx.
The behaviour associated with śubha-upayoga is listed at Pravacanasāra 3:46ff. This consists mostly of devotion and service to (more) advanced monks and to the ascetic community as a whole. Gāthā 3:54 concludes:
This conduct is commended for ascetics, but it is said to be the best (or 'the highest form of conduct') for householders; by it alone he (the householder) attains highest bliss.
This is interesting because it seems to hint at the possibility of two routes to the goal of final liberation, associated with the two social and religious roles of ascetic
135 On correspondences between this and lay vrata, see Faddegon p. 45 fn. 1. But whether it fits into a specific technical pattern or not, the gāthā clearly describes good lay behaviour.

Page #242

228 Harmless Souls and householder / layman. Perhaps the Hindu idea (most famously expressed in the Bhagavadgitā), that the route to salvation is through the conscientious performance of one's svadharma, is at work here; more precisely, it may be a concession to expectations within the Jaina lay community which have been aroused by the realisation that, ultimately, their Hindu neighbours expected to be rewarded simply through pursuing their dharma as householders.
This interpretation relies, of course, upon taking param sokkham to mean complete liberation and not some less final state. The Tattvadīpikā on 3:54 has no doubt: it explains that good conduct, which is the form of śubhaupayoga, is secondary (gauna) for śramaņas, but for the laity it is primary (mukhya), and even though there is still the existence of passion (rāga), the pure self (suddhātman) can be experienced like the sun's heat experienced through the medium of crystal. Such conduct 'gradually brings about the highest happiness of nirvāṇa'.136 In other words, for Amộtacandra, there is apparently a quick (but difficult) and a gradual (but easier) route to liberation, although it is not clear how gradual the latter is - i.e. whether or not it stretches over many re-births. Whatever the answer, this gāthā clearly reflects a different and more positive attitude to the laity than that evidenced in Book 1 of the Pravacanasāra. This suggests that the material in Books 1 and 2, on the one hand, and that in Book 3, on the other, may have been collated under different social circumstances and with a different audience in mind (not to mention by a different hand).137.
136 Faddegon's trans., p. 190, of kramataḥ paramanirvāņasaukhyakāranatvāt.
137 Note 2:97 - cited above, p. 134, - where the teaching of the niścaya view (i.e., the view which deals with the real nature of t is directed exclusively at śramanas and yatis. See also Pravac. 2:102 and TD, and Faddegon's trans., p. 147, fn. 1. Jayasena's commentary

Page #243

Kundakunda: The Pravacanasāra 229 All this points to the vulnerability of ascetic territory to invasion by the laity once practice has been largely internalised. The upayoga doctrine, with its corollary of internal cultivation, carries within itself the possibility that the condition of suddha-upayoga, and thus of liberation, may be reached by inner development alone. In other words, the attainment of the soteriological goal does not logically entail becoming a śramaņa. Indeed, at the end of the Pravacanasāra, in gāthā 3:74, there is what looks like a re-definition of 'śramana' in terms which would not necessarily exclude the laity. Gāthā 3:74 reads:
He, who is pure, is said to be a śramaņa; to the pure one belong faith and knowledge; the pure one attains liberation; he alone is a siddha: my salutation to him. 138
In other words - although perhaps no Jaina would want to put it as bluntly as this -, if you fulfil these criteria through inner-discipline, you are a 'śramaņa' regardless of your external, social status.
(Tātparyavrtti) on this gāthā gives a second interpretation of the Prākrit sāgāro nägāro as sāgārānāgāraḥ, 'while he is a householder or an ascetic', as an alternative to the first interpretation, viz. sākārānākāraḥ, 'with formed or with formless (self-realisation)'. If Jayasena's suggestion were correct, this would mean that meditation on the pure self was considered liberating for both ascetics and laymen. This seems unlikely to be the meaning in the context, but the fact that such a reading could be considered indicates, if nothing else, the status and content of lay religious practice at the time the Tātparyavrtti was composed (second half of the 12th century C.E.? - see Upadhye p. civ).
138 Upadhye's trans., p. 34. The TD comments: 'śramaņa-hood, which is the manifest road to liberation and is characterised as mentalconcentration occupied with a simultaneity of perfect conviction, knowledge and conduct, belongs to the pure alone' - Faddegon's trans., p. 198 of : yattāvat samyagdarśanajñānacāritrayaugapadyapravrttaikāgryalaksanam sākṣān mokşamārgabhūtam śrāmanyam tac ca śuddhasyaiva.

Page #244

230 Harmless Souls
However, as we have seen, even within the Pravacanasāra itself there are indications of built-in checks to hold back any serious claim on the part of the laity to inner purity for themselves. Put simply, the only evidence of internal purity is external behaviour, and the latter is defined in śramaņic terms. We may speculate that, without this check, a fully internalised and laicised Jaina religious practice, based on what is essentially a pan-Indian meditational technique, would have been likely to prove critical for the cohesion and identity of the Jaina community. Moreover, without the continuing necessity of strict physical (i.e. external) ahimsā at some level - for whatever reason - there would have been nothing to guarantee the most obvious 'emblem' of Jaina religion or, we may suppose, to retard the tendency towards ethical decay.
I shall now turn to the Samayasāra, which provides a test case as to whether Kundakunda can, or even wants to, sustain this tenuous link with external practice in what is his least orthodox work.

Page #245

PART IV KUNDAKUNDA: THE SAMAYASĀRA
Kundakunda: definitions and truths
7.1 Samayasāra
i) Introduction Turning from the Pravacanasāra to the Samayasāra, we soon become aware of a number of significant differences in terms of content and scope between the two texs. Some of these differences are so great that it is difficult to think that the two works, or even particular parts of them, should be ascribed to the same author (or redactor). (This, of course, does not affect my argument, which is concerned with the practical and historical implications of certain trends in Jaina doctrine.) Three differences are particularly significant. First, there is the absence in the Samayasāra of any upayoga doctrine. As we have seen, this was Kundakunda's (i.e. the author-redactor's) distinctive means of explaining the mechanism of bondage in the Pravacanasāra. The term upayoga does occur in the Samayasāra, but only in the restricted technical sense employed in, for instance, the Tattvārtha Sūtra. Jayasena, in his Tātparyavrtti commentary on the Samayasāra, does occasionally employ the Pravacanasāra-type upayoga doctrine for exegetical purposes (e.g. at Samayasāra 210), whereas Amstacandra, in his Ātmakhyāti is apparently more attached to the immediate text. There is, however, an incompletely formulated bhāva doctrine in the Samayasāra which fulfils a function similar or identical to that of the upayoga doctrine. (This will be considered below.)
The second significant difference is one of degree
1 See above, 4.2(ii): Upayoga according to the Pravacanasāra.

Page #246

232 Harmless Souls
rather than kind. The use to which the niscaya-vyavahāra doctrine is put in the Samayasära is, in a number of gāthās, far more explicitly radical than anywhere in the Pravacanasāra, and the implications for ethical conduct are consequently more serious. In fact the niscaya view is applied in two different and incompatible ways in the Samayasara.
The bifurcation of the niscaya doctrine is probably connected with the third significant difference between the two texts: the fact that the Samayasara is not only more obviously concerned with confronting and refuting other doctrinal positions, particularly Buddhist and Sāmkhya ideas, but is also more clearly influenced, both in its technical terminology and in its approach to particular problems, by non-Jaina doctrines. It also has a more persistent devotional strain than does the Pravacanasāra; of particular note is the conjunction of the (lay) vocabulary of bhakti with ascetic concentration on the self (in 'selfdevotion') as a means to liberation. (This was, perhaps, an attempt to reduce, through assimilation, the attraction of the Hindu bhakti cults for Jains.)
The relative eclecticism of the Samayasara thus indicates the probability that it was compiled from a number of sources, each of which had been subject to a variety of influences. It is not my purpose here to tease out all these threads (although this is an area in which more research could be fruitfully conducted); rather I intend to illustrate the ways in which the Samayasara represents a point of maximum tension between theoretical philosophy and the Jaina tradition of ascetic practice. I shall then consider how certain Jaina philosophical strategies - such as anekāntavāda (the doctrine of manifold aspects) and syādvāda (the doctrine of qualified assertion) - are used alongside Kundakunda's doctrine of 'two truths' in an attempt to hold together these two strands, the theoretical and the practical.
I shall begin with a general discussion of how the text

Page #247

Kundakunda: The Samayasāra 233 views itself, i.e. of how it defines the samaya of which it is the essence (sāra).
ii) Samaya Perhaps the most obvious translation of samaya would be 'rule' or 'way of life', and thus the Samayasāra would be 'the essence of (our) way of life'.2 However, following Amstacandra and Jayasena, modern commentators take samaya to mean the realised or unified self. At first sight this seems puzzling. How have they arrived at this apparently idiosyncratic definition?
Without prior knowledge of the way in which the term should be understood all the internal evidence is inconclusive. An external source, however, points us in the right direction. Vātsyāyana's Nyāyabhāşya (dated by Frauwallner to the first half of the fifth century)3 defines samaya as follows:
The direct meaning of this word (samaya) is this referent: this is the application of the rule which connects the designator and the designated. When it is applied the correct understanding of the meaning is derived from the word.4
2 The underlying meaning here is 'what is mutually agreed' - so samaya would be rules for behaviour rather than doctrinal laws. In Tantra it has the standard meaning of 'the way to behave'. Caillat (1987, p. 508) translates Samayasāra as 'Essence of the Doctrine'. Friedhelm Hardy, in a personal communication, points out that, in southern India, 'religion' or 'true religion' developed as the most popular meaning of samaya. (See the Tamil Lexicon, Vol. 3, pp.12916 - 936, under 'camaya'.) I am not, however, persuaded that it should be translated in this way in all or most cases in the Samayasāra, for the reasons given in the following pages. 3 Frauwallner 1973, Pt. 2, p. 8.
asya śabdasya idam arthajātam abhidheyam ity abhidhānābhidheyaniyamaniyogah / tasminn upayukte sabdad arthasampratyayo bhavati -
Nyāyabhāsya. 2.1.55. The Nyāyakośa also gives nirdeśa - 'description', 'specification' - as a synonym for samaya.

Page #248

234 Harmless Souls
In other words, samaya means 'the correct or true definition'. Applying this to the Samayasāra, it quickly becomes clear that there too samaya has the primary or underlying sense of 'true definition'. Thus gāthā 2 of the Samayasāra reads:
Know that the term 'jiva' when it has reference to (right) conduct, faith and knowledge is stringently defined; know that when it has reference to material karma it is loosely defined i.e. it includes that which is essentially alien to it). S
There is an analogous gāthā in the Pañcāstikäya (162):
The jīva is defined by / in its own nature (sahāva / sva-bhāva); when it has inessential (non-defining) qualities for its modes it is alienly (loosely) defined (parasamao / parasmaya) [i.e. it is being 'defined' by what is essentially other than itself]. If one applies the self's own definition one will escape from the bondage of karma.
Returning to the Samayasāra, gāthā 3 reads:
The definition which determines its (the jīva's) unity is universally fine (correct), so talk of bondage when there is only one thing is contradictory.
If what is being referred to here is the astringent definition (sva-samaya) of the self then this gāthā appears to contradict the previous one [2], where three things (cāritra, darśana, and jñāna) are mentioned. However, gāthā 7 resolves the difficulty. There it is stated that:
From the conventional point of view conduct, faith and knowledge are predicated of the knower; but there is neither knowledge,
5 Upadhye (p. xlv) gives sva-samaya as 'the realisation of the self as identical with Right faith, etc.' and para-samaya as identification of the self 'with material karmas'; but this is a gloss rather than a translation.

Page #249

Kundakunda: The Samayasāra 235 conduct nor belief, just a pure knower.
As we have seen, (sva)samaya is the self defined according to its own nature; that is to say, it is defined as one thing, pure knowledge, there being no difference in this context between knowledge and the knower. That is what constitutes the soul's svabhāva or essence, as Samayasāra 198 makes clear:
Various types of rising and fruition of karmas have been described by the great Jina(s); but they are not my own-nature. I am one, a knower by nature.?
That is to say, the unified self - the self which has realised its own nature - is the pure knower (i.e. the knower par excellence, the omniscient, liberated self). And for the pure, liberated knower there is no ratnatraya (samyag jñāna, darśana and cāritra) [gāthā 7].
The heterogeneity of the Samayasāra has already been commented upon; nevertheless, it is possible to chart the ways in which the meaning of the term samaya is extended within the text. As we have seen, the basic meaning of samaya is 'true / correct definition'. Thus the sva-samaya is the astringent definition of the self, i.e the soul defined or described from the perspective of its essence. This is also precisely the condition of the liberated self. From this basis, the term samaya then becomes short-hand for the principal thing it is defining in the Samayasāra, namely, the essential or realised self. In other words, we move from the statement that 'The samaya (the true definition) of the jiva is the realised self to the understanding that 'Samaya is the realised self. Gāthā 151 and its commentaries provide us
6 Cf. Samayasāra 16 (18).
7 Cf. Samayasāra 38 where the self is described as follows: 'I am one, indeed, pure, consisting of faith and knowledge (i.e. upayoga), always incorporeal. Nothing other is mine whatsoever, not even an atom.' See also Samayasāra 31.

Page #250

236 Harmless Souls with a clear instance of this. The verse reads:
The true meaning, indeed, the samaya, pure, sheer, the seer, the knower - the mendicants who are absorbed in this, their own nature, attain nirvāņa.
These epithets obviously all refer to the pure self, so samaya here has become a synonym for the 'unified' or 'realised self'. (Amrtacandra glosses it in terms of 'entering into the knowledge which is produced from the state of oneness'. Jayasena says he is called samaya because 'he attains / transforms (himself) into pure qualities and modes'.): There is also a self-conscious yet ambiguous attempt within the text to define what is meant by 'samayasāra' itself. Gāthā 142 states:
Karma is bound or not bound to the self - know these to be points of view. But whatever is said when these alternatives have been transcended, that is samayasāra (the true definition of jiva).
Is Samayasāra therefore a teaching - a 'view' which transcends views? Or is it a condition of the self which reflects its true, transcendental relation to karma?9 The Ātmakhyāti is almost equally ambiguous, stating that he who goes beyond both views and their combination acquires or finds samayasāra. And 'if that is the case, then who, indeed, would not activate the alternative-renouncing state of mind?'10 The fact that alternatives are being definitively discarded lends some weight to the idea that samayasāra (essential definition) cannot be another view,
8 ekibhāvapravsttajñānagamana - Ātmakhyāti JGM ed. on Samayasāra 161 (= Chakra. 151). śuddhagunaparyāyān pariņamati - Tātparyavrtti on ibid.
9 I shall deal with this question from a different perspective when I consider Kundakunda's use of the vyavahāra-niscaya doctrine. See below, pp. 239ff.
10 yady evam tarhi ko hi nāma pakşasamnyāsabhāvanām na nātayati - Ātmakhyāti on Samayasära 142 (152 JGM ed.).

Page #251

Kundakunda: The Samayasāra 237 however transcendent; nevertheless, the paradox of a 'definition' that transcends all views cannot be totally discounted.
The text continues:
One connected to the true definition (samaya) knows what is said of the two views, but does no more; he takes neither of the alternative views at all, being without (such) alternative views. [Samayasāra 123]
But again samaya could be translated here either with the primary sense of 'true definition' or with the extended sense of the realised or unified self. (It should be noted, however, that this gāthā is capable of providing an accurate description of the pure omniscient self that just knows.) This is also true of Samayasāra 144, upon which the Ātmakhyāti comments:
Whatever is the practice of the cessation of all views through the nonexperience of any point of view, that, indeed, is samayasāra.11
Modern commentators, however, seem to be in no doubt about the principal meaning of 'samayasāra'. Chakravarti writes:
The term samayasära means the essential nature of the Self. This Absolute Ultimate Unity is transcendental in nature. Hence the various appellation based upon different points of view really have no relevancy in that state.12
J.L. Jaini concurs: 'Samayasāra is the pure soul in its essence'. The soul is really above all impurities, and 'one
11 yah khalv akhilanayapakşākṣunnatayā viśrāntasamastavikalpavyāpāraḥ sa samayasāraḥ - Ātmakhyāti on 144 (154). I shall have more to say about these passages and their relation to the rest of the Samayasāra in my discussion of the 'two truths' doctrine.
12 Commentary on Samayasāra 144, p. 101.

Page #252

238 Harmless Souls
who meditates upon this again and again acquires selfabsorption, which is Samayasāra'.13 Singh remarks that, in the Samayasara, Kundakunda 'gives instructions how to know the real self (Sva-samaya). This Sva-samaya (Samayasara 2) or the Ego-in- itself is the pure and ultimate reality'.14 This is the self which has realised its oneness, a description which 'very much resembles that of the Upanisadic and Advaitic Brahman or Atman'.15
In so far as these definitions reflect the principal teaching of the text they are indeed correct. For we have seen how the weight of meaning of the term samaya shifts from 'definition', via 'the correct definition of the self, to become synonymous with the self in its true nature, i.e. 'the realised or essential self. However, the crucial nuance in this conflation of epistemology and ontology is the one which makes knowledge of the true nature (definition) of the self instrumental in realising that pure self, i.e. instrumental in, or indeed tantamount to, liberation. (I shall have more to say about this below.)
Referring to Upanisadic doctrine, Gombrich (quoting Malamoud) remarks that, 'The identification of one's ātman and brahman is "at the same time the truth to be discovered and the end to be attained"".16 Rephrasing this formulation to fit Jaina circumstances, we may say that the realisation of the true nature of the self- as totally separate from and untouched by the other (karma, etc.) is at the same time the truth to be discovered and the end to be attained. Consequently, there are passages such as that at Pañcāstikāya 162 17 which state that the realisation of the correct definition of the self is actually the means to liberation. In other words, the definition or description of the soul from the perspective of liberation of the soul as it
13 Commentary on 151 (= 144 Chakravarti's ed.).
14 Singh p. 85.
15
Ibid. P. 89.
16 Gombrich 1988, p. 43.
17 Quoted above, p. 234.
-

Page #253

Kundakunda: The Samayasara 239
is in its self (svabhāva) - is conflated with the idea that that state of the liberated self is actually achieved by realising, i.e. 'knowing', that it is the true or real state of the self. And it is in this way that the realisation or 'definition' of the self is seen to be both means and end, the way to liberation and the state achieved.
So whether samaya is read as the true condition of the self or as a definition or 'view' beyond alternatives, it comes finally to imply self-realisation and thus liberation. This process in itself illustrates the growing stress on 'selfrealisation' in Jaina doctrine, even if some of the gāthās collected under the heading of Samayasāra were not originally so gnostic in tenor. It is the gnostic, however, which in the end comes to overlay and alter the meaning of the other layers. A change of context, with its new juxtapositions, inclines some older or more orthodox doctrines towards new meanings. Whether or not one attributes all these shades of meaning to an individual compiler ('Kundakunda') is strictly not relevant to my purpose, which is to explicate a particular doctrinal tendency and its implications for practice.
7.2 Vyavahāra-niscaya: the two truths doctrine
We have already seen how the doctrine of two truths was employed in the Pravacanasara. The use to which this vyavahāra-niścaya doctrine is put in the Samayasāra is more complex and requires some independent discussion before we examine the ways in which it is applied to the mechanism of bondage and liberation.
First, I shall point out a number of apparent contradictions in the text. Gāthā 8 reads:
Just as a non-Aryan is not able to make another understand [anything] without his non-Aryan speech, so without the conventional truth instruction in the highest truth is not possible.
That is to say, initial instruction has to be couched in

Page #254

240 Harmless Souls terms which are capable of being understood by those who are spiritually undeveloped. This is the standard reconciliation of the elements in a two truths doctrine: the lower is a means or ladder to the higher. As Frauwallner puts it:
The common-place consideration (vyavahāra-nayah) is necessary in order to make the doctrine intelligible to ordinary men. One can only come to an understanding with a foreigner when one uses his speech (Samayasāro (sic.) v.8]. But one must be clear about the fact that it has validity only in a certain sense. It should necessarily supplement the pure way of thought which alone brings full truth.18
This relation is typified in gāthā 16 (19):
(Right] belief, knowledge and conduct should always be practised by a sādhu (from the vyavahāra point of view]; but know that these three are, in reality, the self.
Amrtacandra comments that it is established that the sādhu, in talking to other people, must refer to belief, knowledge and conduct. Personally, however, he must cultivate the self with the sentiment that it is at the same time both the means and the end.19 In other words, the sādhu sees through the vyavahāra view.
Referring to one of the last gāthās of the Samayasāra, Amrtacandra also states that those who are 'deluded by beginninglessly produced vyavahāra views ... do not see the highest truth, the holy Samayasāra'.20
Similarly, the Samayasāra states:
18 Frauwallner 1973, Vol. 2, p. 208.
19 yenaiva hi bhāvenātmā sādhyam sādhanam ca syāt tenaivāyam nityam upāsya iti svayamākūya pareșām vyavahāreņa sādhunā darśanajñānacāritrāņi nityam upāsyānīti pratipadyate - Ātmakhyāti on Samayasāra 16 (= 19 JGM).
20 te 'nādirudhavyavahāravimūdhāḥ ... paramārthasatyam bhagavantam samayasāram na paśyanti - Ātmakh on Sam 413 (= 443).

Page #255

Kundakunda: The Samayasara 241
The knowers of reality say that other substance is 'mine' in conventional parlance, but they know that, from the niscaya point of view, not even an atom is 'mine' [324]. Just as when a man says that 'the village, country, town, kingdom are mine' when they are not [really] his, it is his self speaking out of delusion [325], so the knower who takes other substance to be his and makes it his own certainly becomes a wrong-believer [326].
are
In other words, there is no simple progression from the vyavahāra view to the niscaya view an understanding of the former does not automatically lead to an understanding of the latter; on the contrary, if Amṛtacandra is to be believed, it actually prevents it.21 This seems to conflict directly with the idea that the vyavahāra in itself is a means to perfecting the self. (And it should be remembered that the elements of standard Jaina metaphysics - the doctrines concerning the various tattvas, padarthas, etc. included under the vyavahāra rubric.) Rather, the crucial step seems to be the recognition that there are two views, and that one of them entails delusion and wrong belief. That is to say, you have to realise that the vyavahāra view is a lower view, that it is just a way of talking about the self for 'practical' or pedagogical purposes, before it can become a means to the higher view. Indeed, it is the recognition of that fact, rather than the doctrinal content of the lower view, which is the real means to achieving the higher, liberating view. It is not surprising, therefore, that the predominant attitude towards the vyavahāra viewpoint in these texts ascribed to Kundakunda should be negative. Gāthās such as Samayasara 8 (see p. 239, above) are rare. More typical is Samayasāra 156:
-

Wise people do not operate in vyavahāra, leaving aside the real object (niccayaṭṭham); the destruction of karma is ordained [only] for those ascetics whose refuge is the highest object [i.e. the pure
21 See Atmakhyāti on 413, quoted above.
-

Page #256

242 Harmless Souls
self, the sole object of the niścaya view).
Similarly, gāthā 166 states that there is no influx of karma (āsrava) or bondage for the right believer (samyagdrsti).22 And as we have seen, the 'right-believer is the person who, rejecting the wrong-belief of the vyavahāra view, sees things from the niścaya point of view.
Seemingly most radical of all, there is Samayasāra 11 (=13]:
vavahāro 'bhūdattho bhūdattho desido du suddanao | bhūdattham assido khalu samăditthi havadi jivo!
The vyavahāra [view] does not deal with the really existent, but the pure view (śuddha-naya) is taught as the really existent. The living being who depends upon the really existent is, indeed, a rightbeliever.
Clearly, the significance of this statement depends to a large degree upon the meaning attributed to bhūdattho (bhūtārtha), translated as 'really existent'. Amrtacandra offers no real definition.23 Jayasena, however, glosses vavahāro as vyavahāranayaḥ, and abhūdattho as abhūtārthaḥ asatyārtho bhavati. Conversely, suddhanao is śuddhanayaḥ niscayanayaḥ and bhūdattho is bhūtārthah satyārthaḥ [Tātparya-vrtti. on Samayasāra 114=13 JGM)]. Thus, while the vyavahāra-naya has what is false as its object, the suddha / niscaya-naya has what is true. In this way, the focus is shifted from the ontological to the epistemological, from things to views about things.
Since the niscaya-naya is the view of the unified self, eternally separate from non-self, and the vyavahāra-naya is the conventional view, which sees the self as interactive
22 ņatthi du āsavabamdho sammāditthissa (nāsti tv āsravobandhaḥ samyagdrșter) - Samayasāra 166.
23 But see below, pp. 250-251 - Puruşārtha quote.

Page #257

Kundakunda: The Samayasāra 243 with the non-self (everything that makes up samsāra), then it is clear that, according to this Samayasāra doctrine, the latter view is 'untrue' in the sense of not reflecting the the real nature of the lack of) relationship between 'self and 'other'. (Bhūtārtha is thus the 'really existent' in the sense of being the truth about the (non-) relationship between ātman and paradravya.)
This seems to be the way in which this gāthā [11] should be taken. However, some modern commentators have read bhūtārtha in a strongly ontological sense - i.e. they believe it to denote that which actually exists. On this reading, abhūtārtha (viz. paradravya or samsāra) is an illusion. Thus Bhatt writes that the contact of jīva and ajīva is a 'pure fiction' (upacāra) and not reality; it brings forth 'illusory experiences constituting the worldly sphere'; consequently, the world is 'a creation of ignorance appearing real only as long as the soul remains ignorant about its true nature'.24 This is tendentious, not merely in the dubious translation of upacāra (a term imported by Bhatt), which might be better rendered as 'metaphor' or 'figure of speech', for Bhatt has jumped from the unreality of a relationship between two categories to the unreality of one of the elements in that relationship (the ajīva). Moreover, he claims that the ajīva is actually a creation of ignorance - a statement clearly made under the influence of a particular interpretation of late Vedānta. It might be possible to argue that Kundakunda is swaying in that direction, but it can hardly be claimed that he calls the reality of the separate ajiva into doubt. Indeed, its reality or unreality is irrelevant to liberation; it is the realisation of its non-relation with the jīva which is crucial. That is to say, it is the relation itself which is the 'creation' of ignorance, not the thing. To make ajīva unreal would be to make bondage itself unreal, the result of an inexplicable delusion rather than the corollary of an understandable confusion of categories. (That the categories are in reality absolutely
24 Bhatt 1974, pp. 279-291.

Page #258

244 Harmless Souls
separate is, of course, a matter of dogma and the premise upon which the argument is based, not part of it.)25
Dixit is closer to the significance of this gāthā [Samayasara 11-13] when he writes of the Samayasāra
that:
the whole of this text is a standing harangue against all talk of a relationship between a soul and a matter (sic.). Towards the very beginning (v.13) we are told that the practical standpoint is the standpoint of untruth while definitive standpoint is the standpoint of truth so that even to concede that from the practical standpoint a soul and matter do enter into mutual relationship amounts to saying that they in fact do nothing of the sort.26
The radical nature of this gāthā [11=13] thus lies in the fact that it states explicitly that the niscaya view is true because it expresses the way things really are, and the vyavahāra untrue because it is a false account of reality. The relativity of truth to viewpoint (syādvāda), based on the manifold (anekānta) nature of reality, seems to have been rejected here in favour of an absolute view of truth. In other words, Kundakunda looks like an ekantavādin here, with a doctrine of 'two truths' which bears a close resemblance to that used in other ekanta systems.
Gāthā 272 [=296 JGM] of the Samayasara is equally explicit:
Know that the vyavahāra view is contradicted by the niscaya view.
25 Here it is useful to bear in mind Matilal's words on 'two truths' in Vedanta and Buddhism, that 'an object can be said to be not real in two very different senses'. It can be non-existent or it can be devoid of the 'own-nature or svabhava that it is supposed to possess or that it professes to possess'. Thus, samsara is 'not a mere appearance, still less an illusion - it is something that is not quite successful in embodying an own-nature, svabhāva' (1986, p. 137). In the terms of the present discussion, it is the self viewed as related to non-self that is not real, because its essence is pure, inactive, isolated consciousness - it is a [self-] knower and nothing else - that is its svabhāva.
26 Dixit 1971, p. 134.

Page #259

Kundakunda: The Samayasāra 245 Munis absorbed in / adopting the niscaya view attain nirvāņa.
The idea of a graduated progression from the vyavahāra to the niścaya viewpoint does not arise here. Liberation is achieved only by rejection of the former and absorption in the latter (i.e. the self). Amộtacandra commments:
The niscaya view refers to the self, the vyavahāra view refers to the other (the not-self]. The conventional view is prohibited by the niścaya view, which blocks for the one who desires liberation all intention which relates to anything other than the self as being a cause of bondage, and because for him intention is no different from being dependent on something other than the self. It is to be rejected because only those are liberated who resort to the niścaya view, which refers to the self, and because it is the abhavya - one who is definitively incapable of being liberated - who resorts to the vyavahāra view which refers to the other.27
So even if the vyavahāra viewpoint were to be considered a necessary first position, there is clearly no natural progression from there to the liberating niscaya view. Indeed, it is difficult to see the vyavahāra view positively at all (i.e. as a ladder or raft to the higher view and liberation) when it is precisely the view held by those who can never achieve liberation, the abhavya souls.28 In other words, the vyavahāra view is essentially a 'wrongview'. (But note that it is not the view itself which prevents the abhavya souls from ever being liberated. Rather that view is their characteristic view; they are incapable of
27 ātmāśrito niscayanayah parāśrito vyavahāranayaḥ | tatraivam niścayanayena parāśritam samastam adhyavasānam bandhahetutvena mumuksoḥ pratiședhayatā vyavahāranaya eva kila pratișiddhaḥ, tasyāpi parāśritatvāvišeșāt| pratiședhya evam cāyam, ātmāśritaniscayanāyaśritānām eva mucyamānatvāt, parāśritavyavahāranayasyaikantenāmucyamānenābhavyenāśriyamānatvāc ca - Ātmakhyāti on Samayasāra 272 (= 296 JGM).
28 On abhavya see JPP p. 140, and P.S. Jaini 1977.

Page #260

246 Harmless Souls holding any other.) It should be abandoned as soon as possible.29
This feeling is expressed in more purely psychological terms at Samayasāra 12 [=14]:
The pure [viewpoint} which teaches about the pure (substance - i.e. the pure self] should be known by [those whose object it is to be] the seers of the supreme mental state; but the vyavahāra teaching is for / employed by those who stand in / employ inferior mental states.
Again the negative formulation indicates that the crucial liberating act is to abandon the vyavahāra view. The śuddhanaya, which should be adopted instead, is defined at Samayasāra 14 [=16):
He who sees the ātman as neither bound nor touched [by karmic matter), not other than itself, fixed, without differences, and not combined (with anything not self], know that he is one who holds the pure point of view.
On this reading, the difference between the two views is irreconcilable: one cancels out or denies the other.
This is quite a different interpretation of the 'two views' or 'two truths' doctrine from that expressed at, for instance, Samayasāra 345-348 (=JGM 357-360]:
From the point of view of modifications the self is destroyed; from another point of view it is not. Because of this, there is not the one-sided view that the soul acts or that something else acts. From the point of view of modifications the self is destroyed; from another point of view it is not. Because of this, there is not the one-sided view that the soul experiences or that something else experiences. . It should be known that whoever holds the doctrine that the self that acts is the self that experiences (the fruits of that action] is a
29 What this entails in terms of conduct will be considered below.

Page #261

Kundakunda: The Samayasāra 247 wrong-believer and not of the Arhat faith. But it should be known that whoever holds the doctrine that the self that acts is other than the self that experiences is a wrongbeliever and not of the Arhat faith.
Here the two views relate to whether an object is viewed with regard to its modes (paryāya) or to its substance (dravya). To take either perspective as the exclusive truth about an object would be one-sided (ekānta) and the mark of a wrong-believer. According to this reading, the self is neither the same as nor different from the doer and experiencer. Right-belief, or 'right-view', entails knowing that both views are valid depending on the perspective taken. Unlike the distinction made above (Samayasāra 14, etc.), this is clearly compatible with anekāntavāda doctrine; one view (i.e. the dravya perspective) is not higher, more 'real', or truer than the other (the paryāya perspective); they are complementary. However, the very fact that the dravya perspective deals with what is fundamental or essential, while the paryāya is a mode of that dravya, indicates at least the potential for a logical hierarchy of truth.
It might be thought that the anekāntavāda-compatible version of the 'two truths' doctrine opens up the possibility of a 'third' view, one which reconciles or synthesises both statements. However, the strictures of the syādvāda doctrine ensure that, rather than a true overview, such a synthesis could only take the form of a perception that the vyavahāra and niścaya perspectives are merely views, and therefore neither is to be taken as exclusively true. Indeed, to ensure this is the function of the classical sapta-bhanginaya formulation as a whole.30
On the other hand, such passages as Samayasāra 141ff. (=151ff. JGM)31 are not readily explicable in syādvāda terms. Indeed, they seem more like an attempt to discard
30 See JPP pp. 94-97; and Dixit 1971 p. 24ff. for the possible historical development of this doctrine.
31 See pp. 236ff. above, and the following.

Page #262

248 Harmless Souls the anekāntavāda perspective entirely. section reads:
The relevant
From the vyavahāra point of view it is said that karma comes into contact with and is bound to the jīva; but from the pure (śuddha) point of view karma neither comes into contact with nor is bound to the jiva. [141] 'Karma is bound or not bound to the jīva' - know these to be [statements made from] different points of view. But it is said that he who goes beyond alternatives [attains) samayasära. [142] One connected to the true definition knows what is said of the two views, but does no more; he takes neither of the alternative views at all, being without (such) alternative views. [143] It is just a statement (vyapadeśa) to say that he [the self] attains right perception and right knowledge; he who is said to be free from any alternative views / viewpoint, he has the essential definition (samayasāra). [144132
The Ātmakhyāti on Samayasāra 141 and 142 (= 151 and 152 JGM) explains:
The vyavahāra point of view is that karma comes into contact with and is bound to the jiva because of the non-existence of any great separateness between them, due to jīva and material karma being modes (paryāya) of a single bound state. The niścaya point of view is that karma does not come into contact with and bind the jiva because of the absolute separateness of jīva and material karma, due to their being fundamentally different substances (dravya). [141] The points of view are twofold: the one view is the conceptualization that karma is indeed bound to the jiva, the other is that karma is not bound to the jiva. Who thus goes beyond, who has altogether overcome conceptualization, having himself become viewless - one whose own-nature has destroyed partial knowledge
32 And so, through the movement from epistemology to ontology outlined above, the person free from alternative views realises, i.e. is, the self as correctly defined.

Page #263

Kundakunda: The Samayasāra 249 - [for him) the essence of self is evident. Therefore, even he who, going beyond the one-sided view that karma is bound to the jiva, conceives that karma is not bound to the jiva does not go beyond views. And even he who, going beyond the one-sided view that karma is not bound to the jiva, conceives that karma is bound to the jīva does not go beyond views. And again, even he who, going beyond these two views, conceives that karma is both bound and not bound to the jiva does not go beyond views. Therefore, he who thus goes beyond the combined view, he alone indeed goes beyond choice. He who thus goes beyond all conceptualization, he indeed finds the essence of self. If that is the case, then who indeed would not activate the view-renouncing state of mind (bhāva)?' [142]33
This is strikingly similar to Mādhyamika Buddhism's denial of the ability of doctrines or views to characterize reality. For the Mādhyamika, according to one modern commentator, 'the real as transcendent to thought can be realised only by the denial of the determinations which systems of philosophy ascribe to it'. 34 In other words,
33 jivapudgalakarmanor ekabandhaparyāyatvena tadativyatirekābhāvāj jive baddhāsprstam karmeti vyavahāranayapakṣaḥ| jīvapudgalakarmanor anekadravyatvenātyantavyatirekāj jive 'baddhasprstam karmeti niscayapakṣaḥ|(141) | yaḥ kila jīve baddham karmeti yaś ca jīve 'baddham karmeti vikalpaḥ sa dvitayāpi hi nayapakṣaḥ| ya evainam atikrāmati sa eva sakalavikalpātikrāntaḥ svayam nirvikalpaikavijñānaghanasvabhāvo bhūtvā sākṣātsamayasāraḥ sambhavati tatra yas tāvaj jīve baddhakarmeti vikalpayati sa jive 'baddham karmeti ekam pakşam atikrāmann api na vikalpam atikrāmati yas tu jive 'baddham karmeti vikalpayati jive baddham karmetyekam pakşam atikrāmann api na vikalpam atikrāmati yah punar jive baddham abaddham ca karmeti vikalpayati sa tu tam dvityam api paksam anatikrāman na vikalpam atikrāmati tato ya eva samastanayapakşam atikrāmati sa eva samastam vikaipam atikrāmati ya eva samastam vikalpam atikrāmati sa eva samayasāram vindati | yady evam tarhi ko hi nāma paksasamnyāsabhāvanām na nātayati (142) | - Atmakhyāti on Samayasāra 141 and 142 (= 151 and 152 JGM)
34 Puligandla p. 289, fn. 100.

Page #264

250 Harmless Souls
'Rejection of all thought categories [concepts] and views [theories] is the rejection of the competence of reason to apprehend reality. The real [according to Madhyamika] is transcendent to thought'.35 'Samayasara' thus looks analogous to that state grasped by Nāgārjuna's prajñā and characterized by paramartha-satya, viz. the direct realisation of a higher, ineffable truth, beyond conceptual thought, which is synonymous with liberation. (It is, of course, only an analogy or, at most, a borrowed or shared technique applied to different material; Buddhist and Jain doctrines concerning the nature of 'self', 'not-self', what constitutes liberation, etc., are, it hardly needs to be said, quite different.)
Amṛtacandra highlights this resemblance in another work, the significantly entitled Puruṣārtha-siddhy-upāya. There he writes:
The niscaya mode (of statement) they describe as having a real referent; the vyavahāra mode as without a real referent. All mundane souls are mostly opposed to knowledge of the reality of things.(5) The great saints (muni) teach the mode without a real referent to wake up the sleepy, (6a) (but) who so understands only the vyavahāra mode, in him there is no teaching.(6b) As to a man who has not known a lion a toy is the only lion, so a man who knows not the real method takes the practical method itself for reality!(7) That disciple alone who understands both the real and the practical method, and takes a higher view equally distinct from both, obtains the full fruit of the teaching.(8)36
35 T.V. Murti, Central Philosophy of Buddhism, p. 208, quoted by Puligandla, ibid. fn. 101.
36 niscayam iha bhūtārtham vyavahāram varṇayanty
abhūtārtham |
bhūtārthabodhavimukhaḥ prāyaḥ sarvo 'pi samsāraḥ ||5|| abudhasya bodhanārtham muniśvarā deśayanty abhūtārtham | vyavahāram eva kevalam avaiti yas tasya deśanā nāsti ||6|| māṇavaka eva simho yatha bhavaty anavagitasimhasya | vyavahāra eva hi tathā niścayatām yāty aniścayajñasya ||7|| vyavaharaniscayau yaḥ prabudhya tattvena bhavati

Page #265

Kundakunda: The Samayasāra 251
First there is the question of what it is the ignorant should understand (6a). Clearly, it is not merely the vyavahāra point of view, since in the second half-verse (6b) that is said not to constitute a teaching at all. Rather, it must be the fact that the vyavahāra mode is just a mode (i.e. that there is a higher view, the niscaya) which should be understood. Beyond this, however, both views must be discarded for a position which is neutral (madhyastha). In other words, the vyavahāra-niścaya distinction is itself simply a means (upāya) to approach the highest goal, a means which must itself be discarded in order to attain final liberation.
Such a view sits uneasily among the orthodox Jaina doctrines of syādvāda and anekāntavāda. For one cannot arrive at such a prescription by adhering to the syādvāda analysis.37 The vyavahāra-niścaya doctrine, however (in those cases where the niscaya view is said to be the truth rather than merely a different perspective), claims that the view that the self is not really bound leads to selfrealisation, a state in which there is no duality of knower and known, and so, by definition, no distinction of viewpoints. The difference between the two views (one congruent with anekāntavāda, the other not) is pointed up by a verse in another work attributed to Kundakunda, the Niyamasāra [159]:
From the conventional point of view the omniscient Lord knows and perceives everything; from the absolute [viewpoint] the omniscient knows and perceives [only] the self.38
madhyasthaḥ |
prāpnoti deśanāyāḥ sa eva phalam vikalam sisyah ||8|| Puruşartha-siddy-upaya, quoted and trans. (with my alterations) by J.L. Jaini 1940, pp. 107-108.
37 There is the idea in the fully developed syādvāda doctrine that the self is avaktavya, but this simply means that 'in some respect (the ontological situation of) the self is inexpressible' - see JPP pp. 95-96. 38 Quoted by P.S. Jaini, JPP p. 267, fn. 33; my translation.

Page #266

252 Harmless Souls
The anekāntavāda-compatible vyavahāra view points to a state of liberation where the self still has some kind of relation with the not-self (there is a 'knower' and something 'known'), whereas the niscaya view envisages a liberated self, totally isolated from the not-self, whose omniscience is identical with self-knowledge.
In other words, even when the Samayasāra is apparently rejecting all views (as at 141-144), and so both anekānta and vyavahāra-niscaya distinctions, it is clear that it is the niscaya view, taken as truth, which leads one to the point where views can be abandoned and liberation achieved, i.e. to self-realization. The niścaya view indeed characterizes - albeit from an inevitably intellectual rather than experiential perspective - the essence of self. Transcending all viewpoints (the intellectual), one goes on to experience the reality of that state. Here, in the Samayasāra, in contrast with the Mādhyamika where the abandonment of views is the chief instrument in the achievement of the goal, it is the espousal of the niścaya view which is clearly of prime importance.39 The idea that the latter should be abandoned too seems to have been added, probably under Mädhyamika influence and perhaps because it was supposed (erroneously) that this was a way of shepherding the absolute, hierarchical distinction of two truths' back into the anekāntavāda fold.40 It may also be the case that what was intended merely as a 'description of 'samayasāra' ('the realized self) in the text itself was extrapolated by the commentators into a technique for achieving that state.
Furthermore, we can see the fundamental incompatibility between syādvāda and the absolute
39 See, for instance, gāthās 11-12, 156, 272, etc., quoted above.
40 The meaning attributed to 'samayasāra' ('the essence of self) in Samayasära 141-144, and the fact that these verses conclude a discrete section of the work, point to the likelihood that they were indeed added after the bulk of the text had already been compiled.

Page #267

Kundakunda: The Samayasara 253
vyavahāra-niścaya distinction. The former is essentially an intellectual strategy for evading confrontation with other schools over what appear to be internal contradictions in Jaina metaphysics, as well as for avoiding the logical resolution of such contradictions, with all that that would imply for ethical conduct.41 The latter, on the other hand, has, like Nāgārjuna's Madhyamika teaching, a soteriological function.
This is not to say that there may not be an awareness at one level of the text of the dangers to Jaina orthopraxy inherent in doctrines present in another layer; the distinction between the gāthās which are compatible with anekāntavāda and those which take an ekanta view is frequently and perhaps deliberately blurred. However, once an absolute niscaya view has been introduced, its superiority from a logical point of view is unambiguous.42 Any attempt thereafter to overlay it with anekāntavāda must be considered unconvincing. The only real solution would be to exclude the absolute niscaya view from the text altogether. The fact that that has not happened only demonstrates its authority within the tradition (whether it derives from an individual called 'Kundakunda', or is merely sanctioned by being attributed to him, is not the point). It therefore falls to the commentators to accommodate it as best they can.
The idea that there are at least two implementations of the vyavahāra-niścaya distinction in the Samayasāra - one compatible with anekāntavāda philosophy, the other (the absolute niscaya distinction) not requires further explication. I shall, therefore, now summarize some of the main ways in which the two patterns are juxtaposed in the text, and suggest some models for their development as contrasting, not to say contradictory, doctrines.
41 The specific nature of the dangers for practice and Jaina identity posed by the absolute vyavahāra-niścaya distinction will be discussed below.
42 See, for example, gāthās 11-12, 156, 272.

Page #268

254 Harmless Souls
7.3 Two 'two truths' doctrines
As we have seen, at least two conflicting patterns of application can be outlined for the 'two truths' distinction in the Samayasāra. According to one of these (I shall call it 'pattern one') the vyavahāra-naya is that viewpoint which considers entities in general, and the jīva in particular, from the perspective of modes (paryāya). The niscaya-naya, on the other hand, is that viewpoint which considers entities from the perspective of pure unified substance (dravya). According to orthodox Jaina doctrine, a substance is that which has qualities and modes .(gunaparyāyavad dravyam).43 The substance is the substratum for the qualities, and the qualities undergo modifications (pariņāma) through acquiring new and losing old modes. 44 P.S. Jaini explains :
Thus, any existent must be seen on three levels: the modes, which last only a moment and belong to the qualities; the qualities, which undergo changes and yet inhere forever in their substances; and the substance, which remains the abiding common ground of support for the qualities and their modes. 45
Any complete description of an entity's nature needs, therefore, to encompass all three levels, and this is what the anekāntvāda purports to do, although at any particular moment 'an ordinary (non-omniscient) person' can only 'be aware of the persistent unity (ekatva) of the substance or the transient multiplicity (anekatva) of its modes'. 46 Thus, from the vyavahāra view, the soul acts and experiences the fruits of action, whereas, from the niscaya view, it neither acts nor experiences fruits; yet neither statement contradicts
43 Tattvārtha Sūtra 5:38. 44 JPP p. 90. 45 Ibid. 46 Ibid.

Page #269

Kundakunda: The Samayasāra 255 the other while they are sheltered under the anekāntvāda umbrella. Both are true from their own particular perspective. There is, therefore, a perspective from which the soul is really bound by karma, for the relation of substance to mode is a real relation.
According to the other 'two truths' pattern ('pattern two'), however, the vyavahāra, or conventional view - that the jiva acts and is subject to the fruits of action - is essentially a 'wrong view'. For the niscaya view, which is 'higher' in the sense of representing the complete truth, not just another aspect of it (it portrays reality), states that, by definition, the jīva can have no connection with ajīva. Thus any perceived relation between the two is nothing more than a delusion, the product of ignorance.
This second pattern is clearly incompatible with the first, i.e. it is incompatible with anekāntvāda philosophy. Moreover, within it the niscaya view contradicts the vyavahāra view, which is not the case in the first pattern. Thus, from the perspective of pattern one, pattern two is an ekānta heresy - a one-sided view; whereas, from the perspective of pattern two, pattern one is in its entirety a 'lower' or vyavahāra view, unrepresentative of the truth.
I am not the first to recognise that the Samayasāra contains a double 'two truths' doctrine. Bansidhar Bhatt has designated every gāthā in the Samayasāra as falling into either a 'mystic' or a 'non-mystic' pattern, the former being attributed by Bhatt to a single individual called 'Kundakunda'.47 However, for reasons which will become clear, I consider this to be too rigid, not to say tendentious, a division of the text. Moreover, Bhatt makes no specific connection between his 'non-mystic pattern (my pattern one) and the anekāntvāda doctrine; and his interpretation of the niscaya view of his 'mystic' pattern (my pattern two) seems to owe more to a wish to claim Kundakunda as a Vedānta metaphysician than to reflect the actual content of the text. I shall, therefore, restrict myself to outlining the
47 Bhatt 1974, pp. 279-291.

Page #270

256 Harmless Souls different conclusions I draw from my own analysis, rather than offering any detailed criticism of Bhatt's thesis.
The difficulties presented to commentators by the juxtaposition within the text of gāthās containing contradictory 'two truths' doctrines should not be underestimated. Perhaps the most acute problem occurs in the attempt to accommodate the second pattern (where the vyavahāra and niscaya views contradict each other) to standard Jaina teachings about bondage. If there is really no contact between the self and karma, how is bondage to be explained? (Underlying this is the even more fundamental problem of Jaina philosophy, viz. how can the immaterial (the self] and the material [karma) ever really be said to be in contact?)
From the perspective of practical soteriology it might be objected that the important thing is the nonidentification of the self with karmic matter and that the ontological status of the two is not strictly relevant. That is to say, whether or not there is actual contact between them, the liberating 'action' is for the self to maintain an attitude of absolute separateness from matter. However, such a claim can hardly be sustained theoretically unless it is also claimed that in reality - i.e. ontologically - there is no relation between them. Liberating knowledge (gnosis) is knowledge of the way things really are; there has to be a correspondence between what is known and what is the case. In other words, transferred to the philosophical or theoretical level (which in the Indian tradition is where they are propagated and defended against criticism) soteriological and ethical doctrines entail ontological or metaphysical counterparts.
The type of problem arising can be illustrated by the following. Gāthā 19 [= 22] of the Samayasāra reads:
So long as there is the understanding 'I am in or I am identical with karmic and quasi-karmic [body] matter, etc. [no-karman],' there is

Page #271

Kundakunda: The Samayasāra 257 [also] lack of true discriminative knowledge. 48
One who knows the true nature of things (bhūdattham / bhūtārtham), however, is not deluded (asammūdho / asammūdhaḥ) in this way [Samayasāra 22 (= 27)]. He knows that upayoga (jñāna and darśana) is the lakṣaṇa of the jīva, so how can such a self become or attribute to itself material substance (pudgala dravya) [Samayasāra 24 (= 30)]?
In Jayasena's recension, two extra gāthās [23 and 24] are inserted after 19 (= 22), clearly with the intention of providing an explanation of how, if the jīva is really separate from ajīva, there can be such a thing as bondage. The second of these verses reads:
Whatever mental state (bhāva) the self produces, he (the self) is the agent (kartā) of that bhāva from the niscaya view. From the vyavahāra view, it (the bhāva) is the agent of material karmas.49
In other words, in reality, the self is responsible for its own mental states (bhāva), and thus its own bondage or liberation (through attachment or non-attachment to the ajiva).50 This is evidently an attempt to distance the jīva from pudgala-karman (the ajiva) as the direct cause of bondage. From the niścaya view the binding state is selfproduced, as opposed to produced by pudgala-karman. However, this does not explain how what is pure consciousness by nature and definition (the ātman) can come to manifest impure mental states (bhāva). As we have seen in the Pravacanasāra, the standard vyavahāraniścaya distinction of the first pattern would have explained
48 On no-karman see the Sarvārthasiddhi on Tattvārtha Sūtra
2:10.
· 49 The grammar here is odd and the text may be corrupt, but I have translated it as it stands with kattāram in line 2 agreeing with bhāvam in line 1 (see Appendix).
50 See Samayasāra (23).

Page #272

258 Harmless Souls that the self is the material cause of its own modifications and the instrumental cause of other modifications, whereas the ajīva (pudgala-karman) is the material cause of its own modifications and the instrumental cause of the jīva's modifications. However, as the commentators are well aware, such an anekāntvāda-congruent explanation, which retains a connection with pudgala-dravya as the instrumental cause of impure bhāvas, is not convincing in the context of the absolute distinction of self and other made by the second vyavahāra-niścaya distinction. Thus, what we have here is an open conflict between the two patterns of the vyavahāra-niscaya doctrine.
Jayasena attempts to resolve this conflict by positing a new distinction - not found in the text itself, but much used by modern commentators 51. between śuddha-niscayanaya, 'the pure niscaya viewpoint' and aśuddha-niscayanaya, 'the impure niscaya viewpoint. So his Tātparyavrttiḥ on Samayasāra 24 (JGM) reads:
From the pure niscaya viewpoint the agent is of pure bhāvas, [while) from the impure niścaya viewpoint (the agent] is of impure bhāvas - so the state of being an agent is just a modification of bhāvas from the niscaya point of view.52
J.L. Jaini explains:
impure thought-activity is attributed to the soul from the impure real standpoint (Ashuddha Nischaya Naya). From the pure real standpoint (Shuddha Nischaya Naya), the doer is the doer only of its own pure modifications.53
This may be clarified further by Niyamasāra 18, which
51 Cf. Uggar Sain trans. and comm. on Niyamasāra 18.
52 nicchayado: aśuddhaniscayanayena aśuddhabhāvānām śuddhaniscayanayena śuddhabhāvānām karteti bhāvānām pariņamanam eva kartytvam | Tātparyavrttiḥ on Samayasara 24.
53 SBJ ed., p. 17 - J.L. Jaini's comments on Samayasāra 24.

Page #273

Kundakunda: The Samayasāra 259
reads:
From the vyavahāra view the self is the agent and experiencer (of the effects) of material karma; but from the niscaya view the self is the agent and experiencer through mental states which have arisen from karma.
This differs from Samayasāra (24) (see above) in that it makes it clear that, according to this niscaya view, particular bhāvas which arise in the self, of which it is the agent or material cause, and which, in turn, are the causes of bondage, only do so through the instrumental influence of pudgala-karman. That is to say, the niścaya view presented here [Niyamasāra 18] corresponds to what Jayasena calls the aśuddha-niścaya-naya. In this way the connection between material karma and the self is kept, but mediated or attenuated through mental states (bhāva). Such a reading clearly belongs to the first vyavahāra-niscaya pattern.
None of this, of course, comes any closer to explaining how. pure consciousness (self) can be subject to impure thought-activity in the first place; all it does do is offer an apparently arbitrary explanation of how the soul, from the niścaya point of view, can be said to be both pure and impure at the same time. In other words, faced in the same text with the two vyavahāra-niscaya patterns outlined above, Jayasena tries to run them together by designating as aśuddha-niscaya-naya what would be considered a vyavahāra view from the perspective of the second pattern. But logically - as perhaps Amrtacandra recognised in avoiding the juxtaposition of these gāthās (Samayasāra 1924 and [23]-[24]) - this makes no difference. The contradictions are unresolved and the problem is left hanging.
Why then did Jayasena feel it so important to attempt to reconcile the two views instead of following what would be a logical solution, that of pattern two - viz. to maintain that in reality the self does not have and cannot have impure

Page #274

260 Harmless Souls bhāvas, that it is not bound, and that it is only necessary to realise the truth of that (non-) relation of the self and karma (the not-self) in order to achieve liberation? Apart from the obvious answer, that he is writing a commentary on a received text and so has to make sense of the contents as he finds them (and he is probably working on the assumption that, despite the disparate nature of the gāthās, he is dealing with a work by the revered ācārya Kundakunda), there is another, more compelling reason. Put simply, because the material covered by the first vyavahāra-niścaya pattern comprises the whole socio-ethical content of Jainism, it is essential for the survival and cohesion of the Jaina community to maintain a connection between that content and the soteriological goal. (I shall return to this below.)
7.4 Dvikriyāvāda
Despite the apparent impossibility of resolving it in orthodox terms, Jaina thinkers were periodically forced to confront directly the problem outlined above - namely, what can be the relation between an immaterial conscious self and material unconscious matter, and how can the latter bind or have contact with the former? One of their strategies for dealing with this involves rejection of the dokiriyāvāda / dvi-kriyāvāda doctrine, the assertion that one cause can produce two different effects in this case, the idea that the soul can be the agent of its bhāvas and also of pudgalakarma or karmic modification). An examination of the way dvikriyāvāda is dealt with in the Samayasāra thus helps to clarify the difference between the two vyavahāraniścaya patterns given above.
Samayasāra 83-86 [=89-92] reads:
From the niscaya view the ātman acts on itself alone. And again know that the ātman experiences itself alone [83]. From the vyavahāra view the ātman acts on various kinds of pudgalakarma. Likewise, it experiences [the fruits of] the various kinds of pudgalakarma (84).

Page #275

Kundakunda: The Samayasāra 277 knowledge
In the light of this equation of knowledge, the knower, and the self, it is revealing to look at Samayasāra 404 (= JGM 434). This is the verse which follows that which equates knowledge and the knower or self (403). It reads:
Knowledge is right belief; self-restraint; the sūtras, consisting of angas and pūrvas; merit and demerit; and asceticism [i.e. pravrajyā - formal assumption of the mahāvrata, and so initiation into mendicancy). 4 The wise agree on this.
There is some confusion here about the form of sammādithi, 'right belief. Chakravarti and Jaini both give sammāditthi 1 samyagdęstih (misprint in Prākrit?, nominative singular in Sanskrit); the JGM edition gives sammāditthi / samyagdrstim (nominative singular Prākrit, accusative singular Sanskrit), which is having it both ways, although it at least conforms to the metrical need for a long syllable. I have translated the verse in accordance with the reading of the commentators and translators, who take jñāna to be the subject. Such a reading certainly conforms to the context. However, it is possible that verse 404 is an imported traditional gāthā whose original subject was samyag-drști, right belief. If this is so, then by the time it was incorporated into the Samayasāra it must already have been very ancient, for the pūrvas had long since been lost.5 Thus if knowledge of the pūrvas is still considered a necessary part of samyag-drsti, which in turn is the prerequisite of the religious life, then, without reinterpretation, this verse is little better than a counsel of despair. Similarly, the idea that full renunciation is necessary for samyag-drsti is very ancient. Therefore, whatever the original meaning of this verse, it is both more
4 See JPP p. 243.
5 If the context were not Jain, however, the obvious interpretation of pūrva(kam) would be simply 'with -', and that may be the meaning here.

Page #276

278 Harmless Souls meaningful and more compatible with the new context to take jñāna as the subject. Indeed, knowledge, which has just been equated with the knower (the self), is clearly too important to the overall significance of the text to revert here to being just another sub-category.
In this context, therefore, Samayasāra 404 states that knowledge - i.e. the knower, the self - is all these things, viz. right belief, restraint, the sūtras, dharma and adharma, and full renunciation. Consequently, to know or realise the self is to know - that is to say, be - all these things. In other words, (self-) knowledge is the whole of the religious life, the mokşa-mārga. Through a radical internalisation a single principle subsumes all the rest, and both the path and the goal are found within the self. Jñāna, which is selfrealisation or gnosis, supersedes all external discipline. (It is also interesting to note how this internalisation removes the problem of the pūrvas: that is to say, their loss is no longer a bar to liberation, since if you know your self you will necessarily 'know all the sūtras - jñāna contains them.) Liberation is thus a function of the autonomous self, i.e. of the cultivation or purification of the self through the development of greater and greater (self-) knowledge. And, as Chakravarti puts it, on perfection of knowledge the self 'becomes perfect and knowledge becomes completely co-extensive with reality'. The self is then both sarvajña and paramātma, the omniscient and absolute self.
ii) Self-realisation We have seen that according to the Samayasāra the self (the knower) and knowledge are in essence one. This is, as it were, the omniscient perspective. But to attain this state, to become the pure self, it is necessary to realise, i.e. to experience, this theoretical truth in the self. In more abstract terms, a description of an ontological condition is both the end of and the means to a new epistemology, the full implementation of which leads back to that ontological
6 Chakravarti p. 233.

Page #277

Kundakunda: The Samayasāra 279 state at a new level of personal experience or attainment. And this liberation (omniscience) through self-realisation or knowledge has been made possible precisely through the equation of self and knowledge. As the Niyamasāra puts it:
From the vyavahāra point of view the omniscient lord sees and knows everything, from the niścaya point of view the omniscient sees and knows the self.7
The clear implication is that by knowing the self one knows everything, and thus one is a liberated soul, a kevalin.
The self to be known is, of course, the pure self, i.e. the self as it is in its essence or svabhāva, the self as knowledge. For 'realising the pure self the jīva becomes pure; but knowing the impure self it becomes impure' [Samayasāra 186).
Indeed, according to the Samayasāra it is only through pure (self-) knowledge that one can attain liberation. The five standard media of knowledge - sense perception, scripture, clairvoyance, telepathy, and omniscience - all refer to the one thing (or state). That is reality (paramatta / paramārtha). Having attained that, one attains liberation. 8 Moreover, 'the many who are without these kinds of knowledge do not achieve this state. If you desire complete liberation (parimokkham) from karma then you must obtain this superior state' [Samayasāra 205].9
The Samayasāra stresses that without concentration, i.e. without at least the attempt to realise this one reality which is the realised self, all religious discipline is useless. Or as Samayasāra 152 (= JGM 162) puts it:
7 Niy 159 (= SBJ 158]. Also quoted pp. 212-213, above. 8 See Samayasāra 204.
9 That paramatho I paramārtha in the Samayasāra refers to the pure, unified or realised self - that it is the condition of 'entering into knowledge which is produced from the state of oneness' - has been established above. See p. 235ff. above, ref. Samayasāra 151.

Page #278

280 Harmless Souls
One who performs tapas and (observes) vrata without being fixed in reality [paramatha - the pure self], the all-knowing call that (practice) foolish tapas and foolish vrata.
So even when they are 'observing vows, rules and restraints, and practising tapas, such people are devoid of knowledge because they are outside paramārtha' [Samayasāra 153). Outside paramārtha, and ignorant of the cause of liberation, they long for merit (punya) which can only keep them in samsāra.10
Knowledge or realisation of the self is thus crucial to liberation, but how is it to be attained? It has already been suggested that partial knowledge is the indispensable means to realisation of absolute knowledge, but what form does such knowledge take?
As we have seen, the ignorant self binds itself through identifying with what is not self. As Samayasāra 92 (= JGM 99] puts it:
The soul (jīva) that is full of ignorance, (mis)taking the non-self (para) for the self (ātman), and the self for what is non-self, is the agent of karmas.
Conversely,
The soul which consists of knowledge, not (mis)taking the non-self for the self, or the self for what is non-self, is not the agent of karmas (Samayasāra 93 = JGM 100].
The essential mental act is therefore discrimination based on knowledge of the true nature of the self. Again we see the mixture of ontology and epistemology: the pure self cannot, by definition, come into contact with anything not self, yet it is precisely knowledge of this which is instrumental in separating self and not-self - i.e. such knowledge causes the self to relinquish its 'unreal' or
10 See Samayasāra 154 (= JGM 164).

Page #279

Kundakunda: The Samayasāra 281 delusory contact with the other and thus realise its real condition.
As we have seen, with increasing stress on the essential purity of the self and its equation with knowledge (omniscience), the idea arises that the ātman, in its essence or svabhāva, cannot by definition ever really come into contact with any material substances. And what was originally true just of the individual who had attained kevala-jñāna is now predicated of every ātman.
This points us towards something which is implied but never made explicit in Kundakunda's reformulation of the relationship between self and matter - namely, that it is impossible that the soul could ever be or ever have been really bound. Naturally, this is closely allied to the shift away from the importance of physical karma and the concomitant rise of the idea that the pure self is something to be attained by realisation of its true ontological status. It should be stressed, however, that because of the eclectic nature of the Samayasāra the direct link with the path of physical austerity is never finally broken. Thus, for instance, while gāthā 71 (= JGM 76) states that:
When the absolute difference between asravas (inflowing karmic matter) and the self (ātman) is known by this jiva, then there is no bondage for it,
the following verse (72 = JGM 77) tempers this with the statement that:
Having known the impurity of asravas, their contrary nature (to the self), and that they are the causes of misery, the jīva abstains from them.
In other words, the jīva that knows the true relationship between the self and paradravya does not act in an asravacausing way and so is not bound. Thus in these verses, it is implied that knowledge is still at least nominally subsidiary to conduct as the means to liberation. Nevertheless, given

Page #280

282 Harmless Souls the overall trend of thought in this text, such qualifications seem little more than a holding operation in a pass which, in principle, has already been sold.
If the state of purity, of omniscience, is the eternal condition of the ātman (i.e. in reality it cannot and can never have been associated with the impure, knowledgerestricting not-self), and the knowledge of this truth realises or brings it about in fact (i.e. engenders liberation), then this comes very close to saying that every individual is now and always has been a kevalin. In other words, there are the same kind of premises here which gave rise in Mahāyāna Buddhism to the tathāgatagarbha or 'BuddhaNature' theory. 11 For, in the Jain case, there is the barely concealed proposition that everyone has the Jina- or kevalin-nature (i.e. each person's ātman is really omniscient), and that knowledge of this realises it ontologically. Kundakunda, of course, does not go so far as to say that samsāra and nirvāņa are really the same thing; it is still assumed that with physical death the liberated ātman is released to a nirvāṇic place or condition. However, the logic of this physical liberation, which was tied to the material view of the universe (and perhaps originally to the idea of a material soul), is now very vague and, significantly, no attempt is made to explicate it. Emphasis now falls upon the state of liberation rather than upon its location. (Again, the concept of liberation as a state of pure consciousness or knowledge is reminiscent of Mahāyānist beliefs.)
One can only speculate on the effect of such ideas on the Jaina layperson, even supposing that they were disseminated to him. But it is clear that any changes must have been largely in terms of expectation rather than practice. That is to say, if the link with physical asceticism cannot be broken without risking the identity of the community as a whole, such theoretical possibilities as that
11 On the tathāgatagarbha, see, for instance, P. Williams 1989, pp. 96-115.

Page #281

Kundakunda: The Samayasāra 283 of a 'Jina-nature' nevertheless make liberation seem closer. They are affectively satisfying. (This, of course, still leaves room for the [heretical] suspicion that practice may not be quite so important as had previously been thought, given that in reality the self is even now liberated.) And one can continue to defer realisation of the inherently liberated and omniscient self alongside an increased hope of ultimately attaining the ideal religious goal.
iii) Discriminative knowledge Returning now to the concern of the ascetic or advanced layperson - i.e. the way to obtain or realise liberation in practice - it is useful to examine in more detail what is meant by discriminative knowledge.
Samayasāra 202 (not found in the JGM ed.) asks:
How can one, not knowing the self (ātman) and not knowing the non-self (anātman), not knowing the jīva and ajīva, be a rightbeliever (samyag-drsti)?
In other words, one cannot even get on to the first rung of the ladder to salvation without such knowledge;12 and yet such knowledge in itself constitutes liberation. This, however, is not an impenetrably vicious circle, as becomes clear if we draw an analogy with early Buddhism's sammā ditthi or 'Right Understanding'. For the Buddhist, Rahula tells us, 'Right Understanding is the understanding of things as they are ... [it] is the highest wisdom which sees the Ultimate Reality'. This understanding is, however, of two kinds:
What we generally call understanding is knowledge, an accumulated memory, an intellectual grasping of a subject according to given data. This is called 'knowing accordingly' (anubodha). It is not very deep. Real deep understanding is called 'penetration' (pativedha), seeing a thing in its trụe nature... This
12 See JPP p. 272 on samyag-drșți as the fourth gunasthāna.

Page #282

284 Harmless Souls
penetration is possible only when the mind is free from all impurities and is fully developed through meditation. 13
For the Jains too, partial, intellectual knowledge puts one on the ladder which leads to absolute knowledge and liberation.
It is interesting to note that, with the developing stress in Jaina theory on knowledge and self-realisation as the means to liberation, there is an increasing tendency to borrow Buddhist terminology; for Buddhism clearly has a more developed and sophisticated ready-made vocabulary for describing inner discipline and experience. One example of this which is significant in the present context occurs at Samayasāra 293-299. The relevant passage reads:
Having clearly known the nature (svabhāva) of what binds and the nature (svabhāva) of the self (atman), he who has no attachment to what binds attains liberation from karma [293].
The jiva and bondage are differentiated [lit. 'cut'] by their own essential and distinctive characteristics. Cut by the knife of discriminative wisdom (panna/prajñā), they fall apart [294]. [When] the jiva and bondage are differentiated by their own essential and distinctive characteristics, bondage should be cut away and the pure self grasped [295].
How is the self grasped? The self is grasped by discriminative wisdom (prajñā). Just as [the self] is separated (from bondage) by prajñā, so it should be grasped by prajñā [296].
The conscious being (cetayita) to be grasped by prajñā, I am that in reality; whatever mental states (bhāva) make up the remainder should be known to be other than mine [297].
The perceiver (daṭṭhā / dṛṣṭā) to be grasped by prajñā, I am that in reality, etc. [298]
The knower (nādā / jñātā) to be grasped by prajñā, I am that in reality, etc. [299].
13
Rahula
P. 49.

Page #283

Kundakunda: The Samayasāra 285 Gāthā 293 makes the usual distinction between the svabhāva of the self and everything else, and relates liberation to non-attachment to, or separation from, the latter. This is the inescapable (i.e. ontological) separation of the jīva and what (apparently) binds it (294a), but it is prajñā which is instrumental in realising that gulf between jīva and ajīva (294b). Such realisation of the pure self is liberation (295).
In other words, separation of the jīva and what binds it is no longer a physical matter - i.e. it is not a matter of burning off karma through tapas and preventing further influx (āsrava) through inaction - it is now an intellectual or mental concern. One realises the true nature of the self through prajñā (discriminative wisdom), and that in itself releases the pure self; it enables it to attain to its true condition.
The use of the term prajñā here is almost certainly borrowed from Nāgārjuna or from Buddhist prajñāpāramitā literature in general. For Nāgārjuna, prajñā is the means by which the higher truth (paramārthasatya) is grasped - it is 'direct, intuitive insight into reality as it is in itself. 14 For Kundakunda, the reality of the self is that it is absolutely unconnected with what is not self, and prajñā isolates, in the first instance, that pure self. It is the initial and crucial intellectual means of discriminating between jīva and ajīva, self and bondage. In this sense, it is synonymous with samyag-drsti (see above). But as in the Buddhist case, prajñā has, according to the Samayasāra, another facet. This is spelt out in gāthā 296 (see above), where it is said that just as the ātman is differentiated (vibhakta) from the not self or from 'bondage' by prajñā, so it is also to be realised or attained (grhītavya) by prajñā. In other words, there are two stages or kinds of prajñā: the intellectual knowledge of the way things are in reality, and the means to the personal realisation or attainment of that reality. This seems to correspond to the early Mahāyānist
14 Puligandla p. 95; see also P. Williams 1989, pp. 42-45.

Page #284

286 Harmless Souls distinction between prajñā which is knowing intellectually, through deep or meditative analysis, the way things must really be, and 'paranormal experience of a meditative absorption directed towards the results of such analysis'. 15
In summary, prajñā is conceived of in two interlinked ways: as deep intellectual analysis, and as something less conceptual, the direct realisation or achievement of that state which has previously been isolated by analysis. The fact that for the Jains the liberated self is omniscient does not conflict with the non-conceptual or non-intellectual form of liberating prajñā (prajñā 'two'), for omniscience cannot be attained by a conscious attempt to expand one's normal limited knowledge. Rather it is only to be achieved by removing the obstacles to one's natural state of omniscience, i.e. by purifying the self, and especially by divesting oneself at a deep level of the delusion that the self is really in any way connected with what is not self. It is the experience of the pure self, rather than simply intellectual knowledge of it, that is liberating.
In contrast to the Buddhist texts, this division between, on the one hand, the intellectual and conceptual, and, on the other, the meditative and non-conceptual, is never systematically analysed or even described by Kundakunda in the Samayasāra. Nevertheless, the use of the term prajñā in the passage given above (especially in gāthā 296) implies precisely such a distinction. And what we saw when we considered dhyāna in the Pravacanasāra 16 is no less evident in the Samayasāra; namely, that liberation is to be attained by meditation on the pure self. For instance, we have already seen in a different context) gāthā 151, where it is said that mendicants who are absorbed in the realised self, their svabhāva, attain nirvāṇa.17 Similarly, gāthās 187-189 state that:
15 P. William's description (1989), pp. 43-44. 16 See p. 201ff., above.
17 tamhi třhidā sahāve muniņo pāvamti ņivvāņam - Samayasāra 151. For the complete verse, see p. 236, above.

Page #285

Kundakunda: The Samayasara 287
The self which, having restrained itself by itself with regard to both meritorious and demeritorious actions, is fixed in perception and knowledge and is free from desire for other things, which, free from all combination [i.e. free from all attachment to what is not self], meditates on the self through the self and disregards karma and nokarma (that) sentient being reflects upon the state of
oneness [187-188].
Meditating on the self, consisting of perception and knowledge and not consisting of anything else [OR 'not thinking (manāḥ) of anything else'], he very quickly realises the self which is completely free from all karma [189].18
In other words, it is meditation on the pure or unified self which is itself instrumental in realising or attaining that pure self. You know or identify the pure self (conceptually) and then you realise (i.e. attain it) by meditating on it. Moreover, meditation on the self comes to be seen as both the acme and index of right conduct. As two gāthās only found in Jayasena's recension of the Samayasara put it:
Indeed, meditation should be practised on knowledge, belief (/ perception) and conduct. But these three are the self; therefore practise meditation on the self [JGM 11].
The ascetic who is constantly engaged in practising this meditation on the self attains liberation from all suffering quickly [JGM 12].
The term used for 'meditation' here is not dhyāna but bhāvanā. In early Jaina texts bhāvanā is connected with the five mahāvrata. It has a range of meanings, from the underlying mental disposition which leads to the right understanding of the vows, to their specific observance. 19
18 On nokarma - quasi-karmic matter which makes up the jiva's bodies, etc. - see Tattvärtha Sutra 2:10.
19 See Schubring paras. 45,167,171; cf. JPP p. 243, fn. 3, and TS
7:3.

Page #286

288 Harmless Souls In the present context, however, the meaning of the term is significantly closer to the standard Buddhist use of bhāvanā, namely, as mental development or mental culture in general and meditation in particular.20
The important point to note here is the stress on meditation on the self as the means to liberation:21 right conduct has been redefined as ātmabhāvanā (meditation on the self), and thus internalised. (There is probably also the implication that if one meditates on the self then one's external conduct is automatically correct.)
iv) Renunciation of bhāva In a similar way renunciation is also internalised: stress now falls upon the renunciation of all (mental) states or bhāva (with the exception, of course, of ātmasvabhāva). Gāthā 34 = JGM 39) reads:
As (self-) knowledge renounces all (mental) states (bhāva), knowing them to be other (than the self), so (self-) knowledge should be considered to be the real / definitive definition of renunciation.
In other words, knowing the self, one recognises and renounces or rejects everything not self, all parabhāva 22 And it is reiterated that knowledge is the svabhāva of the self:
The holy men, who know absolute reality, call that holy man a conqueror of delusion who, having overcome delusion, realises that the self has knowledge for its own-nature (svabhāva).
Seen correctly, therefore, the self, untouched by anything else, is in a natural state of renunciation. It does not have to do anything to renounce since, having no other
20 See CPD p. 36, entry on bhāvanā, and Rahula p.68. 21 Cf. Samayasāra 151, above. 22 See Samayasāra 35 (= JGM 40).

Page #287

Kundakunda: The Samayasāra 289 states and being in contact with no other states, there is nothing to renounce. Again knowledge of one's true nature leads to attainment of that true nature; renunciation of parabhāva - i.e. the realisation that they are other (para) - leads to the state of renunciation. For the delusion (moha) which is overcome is precisely that the self has anything to do with other substances. This is made clear by gāthās 36 and 37 (JGM 41 and 42). The first of these states that:
(When) it is realised that 'delusion has nothing whatsoever to do with me; I am one, consciousness (upayoga)', the knowers of the true definition of self (samaya) call that the state of being free from delusion.
The second gāthā (37) repeats this formulation with dharma etc.' (dhammādi / dharmādi) substituted for moha.
In other words, the self, when truly defined, has nothing to do with the other dravya (viz. dharma, adharma, pudgala, akāśa, kāla, and other jīva), and to realise that is to be free from delusion.23
That the knower, the pure self, is naturally a renunciate is borne out by a further passage in the Samayasāra. Gāthā 210-214 read:
Non-possession (aparigraha) is said to be desirelessness. The knower does not desire merit (dharma). And it is by the nonpossession of (i.e. the lack of desire for) dharma that he becomes / is a knower [210 = JGM 225].
The next three gāthās [211-213 = JGM 226, 228-229] repeat this formulation, substituting demerit (adharma) [211], and food (asana) [212] and drink (pāna) (213] for dharma.
The passage continues:
23 At least, Amrtacandra interprets dharma here as the principle of motion, one of the five ajīva substances - see Ātmakhyāti on JGM 42. It could, however, simply mean 'merit'.

Page #288

290 Harmless Souls
Thus the knower does not desire all these various states (bhāva), for in reality his nature (bhāva) is knowledge, and he is independent of everything [214 = JGM 230).
As we have already seen, both meritorious and demeritorious actions keep one in samsāra. Here, the desire for the fruits of such actions has the same result. More interesting in this context, however, is the internalisation of aparigraha with regard to eating and drinking. Now it is the attitude, the lack of desire for food and water, which defines fasting. The stress has been shifted from the material fact of non-consumption, so closely tied to physical ahimsā, to the underlying state, or mental attitude. And aparigraha itself is defined as desirelessness, an attitude.
Thus the indices of the religious life - physical renunciation and non-possession (aparigraha) - are, through internalisation, drawn into the equation of selfknowledge and liberation. And the condition of renunciation is no longer something to be achieved through action or inaction but, being the natural state of the self, it is therefore something to be realised. In other words, the barrier to full renunciation and thus liberation is, in effect, a delusion with regard to what is not self: the delusion that it is possible in reality to have a relationship with it. Consequently, it is not the physical objects which make up the ajīva world themselves which are to be renounced, but the attitude towards those objects. As Samayasāra 210 puts it, 'a person who has no parigraha is said to be desireless' [apariggaho aņiccho bhanido); and it is by this kind of non-possession or desirelessness that one is a knower, i.e. liberated.24 The concomitant of this is that it is attachment or desire, an attitude, which prevents selfknowledge and thus liberation. So according to Samayasāra 20 (= JGM 214):
24 See p. 289, above.

Page #289

Kundakunda: The Samayasāra
291
If as much as an atom of desire / attachment (rāga), etc., is found in someone, he does not know the self even if he knows all the scriptures.
Absence of desire (rāga) is the indicator of selfknowledge because only someone who knows the self, and thus the real relation of self to not self, knows that to desire is pointless, since in reality it is impossible to possess anything at all. In fact this works both ways: desire obstructs omniscience and liberation, but (partial) knowledge leads to the abandonment of desire and thus ultimately to liberation (total knowledge).
How does this treatment of desire (rāga) relate to the standard kaşāya doctrine as found in the Tattvārtha Sūtra ? According to the kaşāya doctrine, it is the negative emotions or passions underlying action which cause karmic particles to adhere to and thus bind the soul. And this seems to be precisely the meaning of a simile employed by Kundakunda at Samayasāra 237-241 (= JGM 255-264). There, it is said of a man with an oil-smeared body, who is performing martial exercises in a dusty place and doing damage to the surrounding foliage, that the real (niscayataḥ) cause of the dust sticking to his body is the oil, not his bodily activity. In the same way, a wrong believer engaged in activity, who has rāga, etc., as his upayoga, is smeared by karmic dust (rajasā) [241 = JGM 259]. On the other hand, when a man performs exercises with a body which is oil-free, whatever the physical destruction he causes, no dust sticks to him. Similarly, a right-believer, even though he is engaged in various activities, is not smeared by karmic dust because of the absence of rāga, etc., as his upayoga [246 = JGM 264].
Chronologically, however, this passage evidently belongs to an early layer of the Samayasāra, given its emphasis on the suppression of passion rather than on jñāna as the means to liberation. And considering the broader context, it becomes clear that Kundakunda's other

Page #290

292 Harmless Souls teachings have entailed significant modifications to the kaşāya doctrine.
At one level a temporary accommodation of the kaşāya and jñāna doctrines is possible: the necessity of controlling the passions is recognised but the means to achieving that control is through knowledge of the true state of the jīva - its fundamental isolation. This leads to the realisation that the desire for a relationship between the self and anything else is bound to be frustrated, and thus to the abandonment of rāga, etc. This accommodation can only be temporary, however, since such a formulation itself contains the doctrinal basis of Kundakunda's second niścaya pattern - the physical isolation of the self from all matter. Thus, karma and the self being totally separate and incompatible, the idea that the self is or can be bound, even by passionate behaviour, is itself a delusion. From this perspective the kaşāya doctrine is intrinsically false.
This illustrates very well the fact that the diverse nature of the material collected in the Samayasāra leads to an ambiguity in the meaning of some individual gāthās which can only be partially resolved by examining them in their immediate context. To take a further example, gāthā 247 (= JGM 265), apparently continuing the passage I have summarised above (Samayasāra 237-246), reads:
He who thinks 'I kill' [Vhims) and 'I am killed by other beings' is deluded and ignorant; but the knower is opposed to (i.e. knows other than this.
Taken as it stands, this seems to belong to the second niscaya pattern - i.e. the self is neither an agent nor a patient but simply a knower. In other words, in so far as they both stress the absolute isolation of individual souls, the difference between this and the Sāmkhya view, expressed famously in the Bhagavadgitā (quoting Katha Upanişad 2:19-20), is nominal. The Bhagavadgitā verse - almost certainly the formal model for Kundakunda's gāthā - states that:

Page #291

Kundakunda: The Samayasāra 277 knowledge.
In the light of this equation of knowledge, the knower, and the self, it is revealing to look at Samayasāra 404 (= JGM 434). This is the verse which follows that which equates knowledge and the knower or self (403). It reads:
Knowledge is right belief, self-restraint; the sūtras, consisting of angas and pūrvas; merit and demerit; and asceticism [i.e. pravrajyā - formal assumption of the mahāvrata, and so initiation into mendicancy]. 4 The wise agree on this.
There is some confusion here about the form of sammāditthi, 'right belief. Chakravarti and Jaini both give sammāditthi | samyagdsstih (misprint in Prākrit?, nominative singular in Sanskrit); the JGM edition gives sammăditthi / samyagdrstim (nominative singular Prākrit, accusative singular Sanskrit), which is having it both ways, although it at least conforms to the metrical need for a long syllable. I have translated the verse in accordance with the reading of the commentators and translators, who take jñāna to be the subject. Such a reading certainly conforms to the context. However, it is possible that verse 404 is an imported traditional gāthā whose original subject was samyag-drsti, right belief. If this is so, then by the time it was incorporated into the Samayasāra it must already have been very ancient, for the pūrvas had long since been lost.5 Thus if knowledge of the pūrvas is still considered a necessary part of samyag-drsți, which in turn is the prerequisite of the religious life, then, without reinterpretation, this verse is little better than a counsel of despair. Similarly, the idea that full renunciation is necessary for samyag-drsţi is very ancient. Therefore, whatever the original meaning of this verse, it is both more
4 See JPP p. 243.
5 If the context were not Jain, however, the obvious interpretation of pūrva (kam) would be simply 'with -', and that may be the meaning here.

Page #292

278 Harmless Souls meaningful and more compatible with the new context to take jñāna as the subject. Indeed, knowledge, which has just been equated with the knower (the self), is clearly too important to the overall significance of the text to revert here to being just another sub-category.
In this context, therefore, Samayasāra 404 states that knowledge - i.e. the knower, the self - is all these things, viz. right belief, restraint, the sūtras, dharma and adharma, and full renunciation. Consequently, to know or realise the self is to know - that is to say, be - all these things. In other words, (self-) knowledge is the whole of the religious life, the mokşa-mārga. Through a radical internalisation a single principle subsumes all the rest, and both the path and the goal are found within the self. Jñāna, which is selfrealisation or gnosis, supersedes all external discipline. (It is also interesting to note how this internalisation removes the problem of the pūrvas: that is to say, their loss is no longer a bar to liberation, since if you know your self you will necessarily 'know all the sūtras - jñāna contains them.) Liberation is thus a function of the autonomous self, i.e. of the cultivation or purification of the self through the development of greater and greater (self-) knowledge. And, as Chakravarti puts it, on perfection of knowledge the self 'becomes perfect and knowledge becomes completely co-extensive with reality'. The self is then both sarvajña and paramātma, the omniscient and absolute self.6
ii) Self-realisation We have seen that according to the Samayasāra the self (the knower) and knowledge are in essence one. This is, as it were, the omniscient perspective. But to attain this state, to become the pure self, it is necessary to realise, i.e. to experience, this theoretical truth in the self. In more abstract terms, a description of an ontological condition is both the end of and the means to a new epistemology, the full implementation of which leads back to that ontological
6 Chakravarti p. 233.

Page #293

Kundakunda: The Samayasāra 279 state at a new level of personal experience or attainment. And this liberation (omniscience) through self-realisation or knowledge has been made possible precisely through the equation of self and knowledge. As the Niyamasāra puts it:
From the vyavahāra point of view the omniscient lord sees and knows everything, from the niscaya point of view the omniscient sees and knows the self.7
The clear implication is that by knowing the self one knows everything, and thus one is a liberated soul, a kevalin.
The self to be known is, of course, the pure self, i.e. the self as it is in its essence or svabhāva, the self as knowledge. For 'realising the pure self the jīva becomes pure; but knowing the impure self it becomes impure' [Samayasāra 186).
Indeed, according to the Samayasāra it is only through pure (self-) knowledge that one can attain liberation. The five standard media of knowledge - sense perception, scripture, clairvoyance, telepathy, and omniscience - all refer to the one thing (or state). That is reality (paramattha
paramārtha). Having attained that, one attains liberation.8 Moreover, the many who are without these kinds of knowledge do not achieve this state. If you desire complete liberation (parimokkham) from karma then you must obtain this superior state' (Samayasāra 205].9
The Samayasāra stresses that without concentration, i.e. without at least the attempt to realise this one reality which is the realised self, all religious discipline is useless. Or as Samayasāra 152 (= JGM 162) puts it:
7 Niy 159 (= SBJ 158). Also quoted pp. 212-213, above. 8 See Samayasāra 204.
9 That paramatho / paramārtha in the Samayasāra refers to the pure, unified or realised self - that it is the condition of 'entering into knowledge which is produced from the state of oneness' - has been established above. See p. 235ff. above, ref. Samayasāra 151.

Page #294

280 Harmless Souls
One who performs tapas and (observes) vrata without being fixed in reality (paramattha - the pure self], the all-knowing call that (practice) foolish tapas and foolish vrata.
So even when they are 'observing vows, rules and restraints, and practising tapas, such people are devoid of knowledge because they are outside paramārtha' [Samayasāra 153]. Outside paramārtha, and ignorant of the cause of liberation, they long for merit (punya) which can only keep them in samsāra.10
Knowledge or realisation of the self is thus crucial to liberation, but how is it to be attained? It has already been suggested that partial knowledge is the indispensable means to realisation of absolute knowledge, but what form does such knowledge take?
As we have seen, the ignorant self binds itself through identifying with what is not self. As Samayasāra 92 (= JGM 99] puts it:
The soul (jīva) that is full of ignorance, (mis)taking the non-self (para) for the self (ātman), and the self for what is non-self, is the agent of karmas.
Conversely,
The soul which consists of knowledge, not (mis)taking the non-self for the self, or the self for what is non-self, is not the agent of karmas (Samayasāra 93 = JGM 100].
The essential mental act is therefore discrimination based on knowledge of the true nature of the self. Again we see the mixture of ontology and epistemology: the pure self cannot, by definition, come into contact with anything not self, yet it is precisely knowledge of this which is instrumental in separating self and not-self - i.e. such knowledge causes the self to relinquish its 'unreal' or
10 See Samayasära 154 (= JGM 164).

Page #295

Kundakunda: The Samayasara 281 delusory contact with the other and thus realise its real condition.
As we have seen, with increasing stress on the essential purity of the self and its equation with knowledge (omniscience), the idea arises that the atman, in its essence or svabhāva, cannot by definition ever really come into contact with any material substances. And what was originally true just of the individual who had attained kevala-jñāna is now predicated of every ātman.
This points us towards something which is implied but never made explicit in Kundakunda's reformulation of the relationship between self and matter - namely, that it is impossible that the soul could ever be or ever have been really bound. Naturally, this is closely allied to the shift away from the importance of physical karma and the concomitant rise of the idea that the pure self is something to be attained by realisation of its true ontological status. It should be stressed, however, that because of the eclectic nature of the Samayasara the direct link with the path of physical austerity is never finally broken. Thus, for instance, while gathā 71 (= JGM 76) states that:
When the absolute difference between äsravas (inflowing karmic matter) and the self (ātman) is known by this jīva, then there is no bondage for it,
the following verse (72 = JGM 77) tempers this with the statement that:
Having known the impurity of asravas, their contrary nature (to the self), and that they are the causes of misery, the jiva abstains from
them.
In other words, the jiva that knows the true relationship between the self and paradravya does not act in an asravacausing way and so is not bound. Thus in these verses, it is implied that knowledge is still at least nominally subsidiary to conduct as the means to liberation. Nevertheless, given

Page #296

282 Harmless Souls the overall trend of thought in this text, such qualifications seem little more than a holding operation in a pass which, in principle, has already been sold.
If the state of purity, of omniscience, is the eternal condition of the atman (i.e. in reality it cannot and can never have been associated with the impure, knowledgerestricting not-self), and the knowledge of this truth realises or brings it about in fact (i.e. engenders liberation), then this comes very close to saying that every individual is now and always has been a kevalin. In other words, there are the same kind of premises here which gave rise in Mahāyāna Buddhism to the tathāgatagarbha or 'BuddhaNature' theory.11 For, in the Jain case, there is the barely concealed proposition that everyone has the Jina- or kevalin-nature (i.e. each person's ātman is really omniscient), and that knowledge of this realises it ontologically. Kundakunda, of course, does not go so far as to say that samsāra and nirvana are really the same thing; it is still assumed that with physical death the liberated ātman is released to a nirvāṇic place or condition. However, the logic of this physical liberation, which was tied to the material view of the universe (and perhaps originally to the idea of a material soul), is now very vague and, significantly, no attempt is made to explicate it. Emphasis now falls upon the state of liberation rather than upon its location. (Again, the concept of liberation as a state of pure consciousness or knowledge is reminiscent of Mahāyānist beliefs.)
One can only speculate on the effect of such ideas on the Jaina layperson, even supposing that they were disseminated to him. But it is clear that any changes must have been largely in terms of expectation rather than practice. That is to say, if the link with physical asceticism cannot be broken without risking the identity of the community as a whole, such theoretical possibilities as that
11 On the tathāgatagarbha, see, for instance, P. Williams 1989, pp. 96-115.

Page #297

Kundakunda: The Samayasāra 283 of a 'Jina-nature' nevertheless make liberation seem closer. They are affectively satisfying. (This, of course, still leaves room for the [heretical] suspicion that practice may not be quite so important as had previously been thought, given that in reality the self is even now liberated.) And one can continue to defer realisation of the inherently liberated and omniscient self alongside an increased hope of ultimately attaining the ideal religious goal.
iii) Discriminative knowledge Returning now to the concern of the ascetic or advanced layperson - i.e. the way to obtain or realise liberation in practice - it is useful to examine in more detail what is meant by discriminative knowledge.
Samayasāra 202 (not found in the JGM ed.) asks:
How can one, not knowing the self (ātman) and not knowing the non-self (anātman), not knowing the jiva and ajīva, be a rightbeliever (samyag-drsti)?
In other words, one cannot even get on to the first rung of the ladder to salvation without such knowledge;12 and yet such knowledge in itself constitutes liberation. This, however, is not an impenetrably vicious circle, as becomes clear if we draw an analogy with early Buddhism's sammā ditthi or 'Right Understanding'. For the Buddhist, Rahula tells us, 'Right Understanding is the understanding of things as they are ... [it] is the highest wisdom which sees the Ultimate Reality'. This understanding is, however, of two kinds:
What we generally call understanding is knowledge, an accumulated memory, an intellectual grasping of a subject according to given data. This is called 'knowing accordingly (anubodha). It is not very deep. Real deep understanding is called 'penetration' (pațivedha), seeing a thing in its trụe nature... This
12 See JPP p. 272 on samyag-drsţi as the fourth gunasthāna.

Page #298

284 Harmless Souls
penetration is possible only when the mind is free from all impurities and is fully developed through meditation. 13
For the Jains too, partial, intellectual knowledge puts one on the ladder which leads to absolute knowledge and liberation.
It is interesting to note that, with the developing stress in Jaina theory on knowledge and self-realisation as the means to liberation, there is an increasing tendency to borrow Buddhist terminology; for Buddhism clearly has a more developed and sophisticated ready-made vocabulary for describing inner discipline and experience. One example of this which is significant in the present context occurs at Samayasāra 293-299. The relevant passage reads:
Having clearly known the nature (svabhāva) of what binds and the nature (svabhāva) of the self (ātman), he who has no attachment to what binds attains liberation from karma [293]. The jiva and bondage are differentiated [lit. 'cut'] by their own essential and distinctive characteristics. Cut by the knife of discriminative wisdom (pannā / prajñā), they fall apart [294]. [When] the jiva and bondage are differentiated by their own essential and distinctive characteristics, bondage should be cut away and the pure self grasped [295]. How is the self grasped? The self is grasped by discriminative wisdom (prajñā). Just as [the self] is separated (from bondage) by prajñā, so it should be grasped by prajñā [296). The conscious being (cetayitā) to be grasped by prajñā, I am that in reality; whatever mental states (bhāva) make up the remainder should be known to be other than mine [297]. The perceiver (datthā / drstā) to be grasped by prajñā, I am that in reality, etc. [298] The knower (nādā / jñātā) to be grasped by prajñā, I am that in reality, etc. [299].
13 Rahula p. 49.

Page #299

Kundakunda: The Samayasāra 285 Gāthā 293 makes the usual distinction between the svabhāva of the self and everything else, and relates liberation to non-attachment to, or separation from, the latter. This is the inescapable (i.e. ontological) separation of the jiva and what (apparently) binds it (294a), but it is prajñā which is instrumental in realising that gulf between jīva and ajīva (294b). Such realisation of the pure self is liberation (295).
In other words, separation of the jiva and what binds it is no longer a physical matter - i.e. it is not a matter of burning off karma through tapas and preventing further influx (āsrava) through inaction - it is now an intellectual or mental concern. One realises the true nature of the self through prajñā (discriminative wisdom), and that in itself releases the pure self; it enables it to attain to its true condition.
The use of the term prajñā here is almost certainly borrowed from Nāgārjuna or from Buddhist prajñāpāramitā literature in general. For Nāgārjuna, prajñā is the means by which the higher truth (paramārthasatya) is grasped - it is 'direct, intuitive insight into reality as it is in itself.14 For Kundakunda, the reality of the self is that it is absolutely unconnected with what is not self, and prajñā isolates, in the first instance, that pure self. It is the initial and crucial intellectual means of discriminating between jīva and ajīva, self and bondage. In this sense, it is synonymous with samyag-drști (see above). But as in the Buddhist case, prajñā has, according to the Samayasāra, another facet. This is spelt out in gāthā 296 (see above), where it is said that just as the ātman is differentiated (vibhakta) from the not self or from 'bondage' by prajñā, so it is also to be realised or attained (gshitavya) by prajñā. In other words, there are two stages or kinds of prajñā: the intellectual knowledge of the way things are in reality, and the means to the personal realisation or attainment of that reality. This seems to correspond to the early Mahāyānist
14 Puligandla p. 95; see also P. Williams 1989, pp. 42-45.

Page #300

286 Harmless Souls distinction between prajñā which is knowing intellectually, through deep or meditative analysis, the way things must really be, and 'paranormal experience of a meditative absorption directed towards the results of such analysis'.15
In summary, prajñā is conceived of in two interlinked ways: as deep intellectual analysis, and as something less conceptual, the direct realisation or achievement of that state which has previously been isolated by analysis. The fact that for the Jains the liberated self is omniscient does not conflict with the non-conceptual or non-intellectual form of liberating prajñā (prajñā 'two'), for omniscience cannot be attained by a conscious attempt to expand one's normal limited knowledge. Rather it is only to be achieved by removing the obstacles to one's natural state of omniscience, i.e. by purifying the self, and especially by divesting oneself at a deep level of the delusion that the self is really in any way connected with what is not self. It is the experience of the pure self, rather than simply intellectual knowledge of it, that is liberating.
In contrast to the Buddhist texts, this division between, on the one hand, the intellectual and conceptual, and, on the other, the meditative and non-conceptual, is never systematically analysed or even described by Kundakunda in the Samayasāra. Nevertheless, the use of the term prajñā in the passage given above (especially in gāthā 296) implies precisely such a distinction. And what we saw when we considered dhyāna in the Pravacanasāra 16 is no less evident in the Samayasāra; namely, that liberation is to be attained by meditation on the pure self. For instance, we have already seen in a different context) gāthā 151, where it is said that mendicants who are absorbed in the realised self, their svabhāva, attain nirvāņa. 17 Similarly, gāthās 187-189 state that:
15 P. William's description (1989), pp. 43-44. 16 See p. 201ff., above.
17 tamhi tthidā sahāve muņiņo pāvamti nivvānam - Samayasāra 151. For the complete verse, see p. 236, above.

Page #301

Kundakunda: The Samayasāra 287
The self which, having restrained itself by itself with regard to both meritorious and demeritorious actions, is fixed in perception and knowledge and is free from desire for other things, which, free from all combination [i.e. free from all attachment to what is not self], meditates on the self through the self and disregards karma and nokarma - (that) sentient being reflects upon the state of oneness [18 Meditating on the self, consisting of perception and knowledge and not consisting of anything else [OR 'not thinking (manāḥ) of anything else'), he very quickly realises the self which is completely free from all karma [189].18
In other words, it is meditation on the pure or unified self which is itself instrumental in realising or attaining that pure self. You know or identify the pure self (conceptually) and then you realise (i.e. attain it) by meditating on it. Moreover, meditation on the self comes to be seen as both the acme and index of right conduct. As two gāthās only found in Jayasena's recension of the Samayasāra put it:
Indeed, meditation should be practised on knowledge, belief (1 perception) and conduct. But these three are the self; therefore practise meditation on the self [JGM 11). The ascetic who is constantly engaged in practising this meditation on the self attains liberation from all suffering quickly [JGM 12].
The term used for 'meditation' here is not dhyāna but bhāvanā. In early Jaina texts bhāvanā is connected with the five mahāvrata. It has a range of meanings, from the underlying mental disposition which leads to the right understanding of the vows, to their specific observance.19
18 On nokarma - quasi-karmic matter which makes up the jīva's bodies, etc. - see Tattvärtha Sūtra 2:10.
19 See Schubring paras. 45,167,171; cf. JPP p. 243, fn. 3, and TS 7:3.

Page #302

288 Harmless Souls In the present context, however, the meaning of the term is significantly closer to the standard Buddhist use of bhāvanā, namely, as mental development or mental culture in general and meditation in particular.20
The important point to note here is the stress on meditation on the self as the means to liberation:21 right conduct has been redefined as ātmabhāvanā (meditation on the self), and thus internalised. (There is probably also the implication that if one meditates on the self then one's external conduct is automatically correct.)
iv) Renunciation of bhāva In a similar way renunciation is also internalised: stress now falls upon the renunciation of all (mental), states or bhāva (with the exception, of course, of ātmasvabhāva). Gāthā 34 (= JGM 39) reads:
As (self-) knowledge renounces all (mental) states (bhāva), knowing them to be other (than the self), so (self-) knowledge should be considered to be the real / definitive definition of renunciation.
In other words, knowing the self, one recognises and renounces or rejects everything not self, all parabhāva.22 And it is reiterated that knowledge is the svabhāva of the self:
The holy men, who know absolute reality, call that holy man a conqueror of delusion who, having overcome delusion, realises that the self has knowledge for its own-nature (svabhāva).
Seen correctly, therefore, the self, untouched by anything else, is in a natural state of renunciation. It does not have to do anything to renounce since, having no other
20 See CPD p. 36, entry on bhāvanā, and Rahula p.68. 21 Cf. Samayasāra 151, above. 22 See Samayasāra 35 (= JGM 40).

Page #303

Kundakunda: The Samayasāra 289 states and being in contact with no other states, there is nothing to renounce. Again knowledge of one's true nature leads to attainment of that true nature; renunciation of parabhāva - i.e. the realisation that they are other (para) - leads to the state of renunciation. For the delusion (moha) which is overcome is precisely that the self has anything to do with other substances. This is made clear by gāthās 36 and 37 (IGM 41 and 42). The first of these states that:
(When) it is realised that 'delusion has nothing whatsoever to do with me; I am one, consciousness (upayoga)', the knowers of the true definition of self (samaya) call that the state of being free from delusion.
The second gāthā (37) repeats this formulation with dharma etc.' (dhammādi / dharmādi) substituted for moha.
In other words, the self, when truly defined, has nothing to do with the other dravya (viz. dharma, adharma, pudgala, akāśa, kāla, and other jiva), and to realise that is to be free from delusion.23
That the knower, the pure self, is naturally a renunciate is borne out by a further passage in the Samayasāra. Gāthā 210-214 read:
Non-possession (aparigraha) is said to be desirelessness. The knower does not desire merit (dharma). And it is by the nonpossession of (i.e. the lack of desire for) dharma that he becomes / is a knower [210 = JGM 225].
The next three gāthās [211-213 = JGM 226, 228-229] repeat this formulation, substituting demerit (adharma) [211], and food (asana) [212] and drink (pāna) [213] for dharma.
The passage continues:
23 At least, Amrtacandra interprets dharma here as the principle of motion, one of the five ajiva substances - see Ātmakhyāti on JGM 42. It could, however, simply mean 'merit'.

Page #304

290 Harmless Souls
Thus the knower does not desire all these various states (bhāva), for in reality his nature (bhāva) is knowledge, and he is independent of everything [214 = JGM 230].
As we have already seen, both meritorious and demeritorious actions keep one in samsāra. Here, the desire for the fruits of such actions has the same result. More interesting in this context, however, is the internalisation of aparigraha with regard to eating and drinking. Now it is the attitude, the lack of desire for food and water, which defines fasting. The stress has been shifted from the material fact of non-consumption, so closely tied to physical ahimsā, to the underlying state, or mental attitude. And aparigraha itself is defined as desirelessness, an attitude.
Thus the indices of the religious life - physical renunciation and non-possession (aparigraha) - are, through internalisation, drawn into the equation of selfknowledge and liberation. And the condition of renunciation is no longer something to be achieved through action or inaction but, being the natural state of the self, it is therefore something to be realised. In other words, the barrier to full renunciation and thus liberation is, in effect, a delusion with regard to what is not self: the delusion that it is possible in reality to have a relationship with it. Consequently, it is not the physical objects which make up the ajiva world themselves which are to be renounced, but the attitude towards those objects. As Samayasāra 210 puts it, 'a person who has no parigraha is said to be desireless' [apariggaho añiccho bhanido); and it is by this kind of non-possession or desirelessness that one is a knower, i.e. liberated.24 The concomitant of this is that it is attachment or desire, an attitude, which prevents selfknowledge and thus liberation. So according to Samayasāra 20 (= JGM 214):
24 See p. 289, above.

Page #305

Kundakunda: The Samayasāra 291
If as much as an atom of desire / attachment (rāga), etc., is found in someone, he does not know the self even if he knows all the scriptures.
Absence of desire (râga) is the indicator of selfknowledge because only someone who knows the self, and thus the real relation of self to not self, knows that to desire is pointless, since in reality it is impossible to possess anything at all. In fact this works both ways: desire obstructs omniscience and liberation, but (partial) knowledge leads to the abandonment of desire and thus ultimately to liberation (total knowledge).
How does this treatment of desire (rāga) relate to the standard kaşāya doctrine as found in the Tattvārtha Sūtra ? According to the kasāya doctrine, it is the negative emotions or passions underlying action which cause karmic particles to adhere to and thus bind the soul. And this seems to be precisely the meaning of a simile employed by Kundakunda at Samayasāra 237-241 (= JGM 255-264). There, it is said of a man with an oil-smeared body, who is performing martial exercises in a dusty place and doing damage to the surrounding foliage, that the real (niścayataḥ) cause of the dust sticking to his body is the oil, not his bodily activity. In the same way, a wrong believer engaged in activity, who has rāga, etc., as his upayoga, is smeared by karmic dust (rajasā) [241 = JGM 259]. On the other hand, when a man performs exercises with a body which is oil-free, whatever the physical destruction he causes, no dust sticks to him. Similarly, a right-believer, even though he is engaged in various activities, is not smeared by karmic dust because of the absence of rāga, etc., as his upayoga [246 = JGM 264].
Chronologically, however, this passage evidently belongs to an early layer of the Samayasāra, given its emphasis on the suppression of passion rather than on jñāna as the means to liberation. And considering the broader context, it becomes clear that Kundakunda's other

Page #306

292 Harmless Souls teachings have entailed significant modifications to the kaşāya doctrine.
At one level a temporary accommodation of the kaşāya and jñāna doctrines is possible: the necessity of controlling the passions is recognised but the means to achieving that control is through knowledge of the true state of the jiva - its fundamental isolation. This leads to the realisation that the desire for a relationship between the self and anything else is bound to be frustrated, and thus to the abandonment of rāga, etc. This accommodation can only be temporary, however, since such a formulation itself contains the doctrinal basis of Kundakunda's second niscaya pattern - the physical isolation of the self from all matter. Thus, karma and the self being totally separate and incompatible, the idea that the self is or can be bound, even by passionate behaviour, is itself a delusion. From this perspective the kaṣāya doctrine is intrinsically false.
This illustrates very well the fact that the diverse nature of the material collected in the Samayasāra leads to an ambiguity in the meaning of some individual gāthās which can only be partially resolved by examining them in their immediate context. To take a further example, gāthā 247 (= JGM 265), apparently continuing the passage I have summarised above (Samayasara 237-246), reads:
He who thinks 'I kill' [Jhims) and 'I am killed by other beings' is deluded and ignorant; but the knower is opposed to (i.e. knows other than) this.
Taken as it stands, this seems to belong to the second niścaya pattern - i.e. the self is neither an agent nor a patient but simply a knower. In other words, in so far as they both stress the absolute isolation of individual souls, the difference between this and the Sāmkhya view, expressed famously in the Bhagavadgitā (quoting Katha Upanişad 2:19-20), is nominal. The Bhagavadgitā verse - almost certainly the formal model for Kundakunda's gāthā - states that:

Page #307

Kundakunda: The Samayasāra 293
He who thinks that it (i.e. the puruṣa) kills and he who thinks that it is killed, neither of them understands; this neither kills nor is killed, 25
However, the next gāthā of the Samayasāra (248 = JGM 266), which immediately ties the connection between death and karma, makes it clear that there is to be no overt abandonment (i.e. transcendence) of ethics. Action is not about to lose its moral significance, and there can, of course, be no karma-transcending theological solution. In context, therefore, gāthā 247 means that what one apparently does or what one has done to one is exclusively the result of one's own karma. Specifically, an individual's death is determined by the expiry of his āyus- or agedetermining karma.26 Moreover, according to Jaina doctrine a person's āyuş-karma (i.e. his longevity) in this life is fixed definitively at some moment during the last third of his previous life.27 In these circumstances, the belief that you are really instrumental in killing others or that they can be really instrumental in killing you is a delusion. The moment of your death has been fixed by karma long before you were (re)born.
More generally (and moving away from the difficulties posed by this quasi-fatalistic doctrine), what happens to individuals is determined solely by their own actions; each is entirely responsible for his or her own karma.28 And it is precisely the deluded belief that one can affect or be affected by others which causes one to be bound by punya
25 ya enam vetti hantāram yaś cainam manyate hatam ||
ubhau tau na vijānīto nāyam hanti na hanyate || Bhagavad
Gītā 2:19 11 26 Samayasāra 248-252 = JGM 266-270.
27 See JPP p. 126. The strong element of fatalism in this has been discussed by P.S. Jaini (1977).
28 See Samayasāra 253ff. = JGM 271ff.

Page #308

294 Harmless Souls and pāpa karma.29 (This does not undercut the ethical imperative of ahimsā because, as we have seen, 30 it is combined with the idea that the determination (adhyavasāna) to kill is enough to bind one (Samayasāra 262, etc.). It does, however, make accidental himsā nonbinding.)
To summarize, the distance between Jainism and Sāmkhya is maintained in orthodox Jaina doctrine (in Umāsvāti, for instance) by the connection between the self and karma. For, at some level, according to the orthodox position, there can be a real relation between the soul and karma, between self and not-self. (Such a conjunction of purusa and prakrti is, of course, an ontological impossibility for the Sāmkhya.) However, when that connection is abandoned as being ultimately untrue (which is the contention of Kundakunda's second niscaya pattern), then the gap between the Jaina and Sāmkhya positions closes. And the verses on karma we have just considered are undercut by statements such as:
Thus the right-believer knows the self, whose own-nature (svabhāva) is to be a knower; knowing reality (tattva), he renounces both the arising and fruition of karma. (Samayasāra 200 (= JGM 212)
In other words, the right-believer realises that karma and the bhāvas it produces have, in reality, nothing to do with his self, so he rejects them. Then, he is instructed:
Having given up the impermanent substances (dravya) and modes of thought (bhāva) in the self, grasp this, your eternal, permanent, single bhāva, realizable through / as your own-nature (svabhāva). (Samayasāra 203 (= JGM 219])31
29 Samayasāra 259-261 = JGM 277-279. 30 See p. 272ff., above. 31 Cf. 297-299 = JGM 325-327, quoted p. 284, above.

Page #309

Kundakunda: The Samayasāra 295 Negatively, therefore, renunciation is the rejection of delusion and wrong attitude (i.e. of alien bhāvas); positively, it is the realisation of the true nature of self, of svabhāva.
Once again, if all the doctrinal elements examined above are directly juxtaposed (i.e. if they are viewed synchronically rather than diachronically) then they are clearly incompatible. The self is not an actor; it is karma which is the real agent. But in reality karma does not and cannot have any effect on the self; therefore, not only is any action by the self a delusion but so too is bondage. That is what one must realise in order to attain liberation. Clearly this threatens both the ethical basis of Jainism and the ascetic practice of physical ahimsā derived from it. In conclusion, therefore, I shall examine the Samayasāra's final verses, which attempt to deal with this question of the significance of ascetic and lay practice.
9.2 Linga, practice and the path to liberation
The Samayasāra concludes with the following eight gāthās:
The deluded, having assumed the characteristic marks of ascetics (pāsamdiya) or householders, which are of many kinds, say: 'This characteristic mark is the path to liberation'. (408) A characteristic mark is certainly not the path to liberation; in that the Arhats who are indifferent to the body, having given up distinguishing characteristics, devote themselves to insight, knowledge and conduct. (409) The characteristic marks belonging to ascetics and householders are not this path to liberation. The Jinas say that insight, knowledge and conduct are the path to liberation. (410) Wherefore, giving up the marks adopted by householders or ascetics, draw the self onto the path to liberation, which consists of insight, knowledge and conduct. (411) Establish the self on the path to liberation, (know it,) meditate on it,

Page #310

296 Harmless Souls
and always dwell there; do not dwell among other substances. (412) The essential self is not known by those who are attached to the various kinds of ascetic or householder characteristic marks. (413) Although the conventional view holds that there are two (kinds of marks) on the path to liberation, by the absolute view one does not wish for any marks on the path to liberation. (414) That conscious being who, having read this Samayaprābhrta and understood its true meaning, holds to that meaning, will attain the highest bliss. (415)
In our examination of Pravacanasāra 3:5 and 3:6,32 we saw that the end of rebirth, and thus the attainment of liberation, is brought about by the adoption of certain pure modes of thought and behaviour. It is these which constitute the defining characteristic (linga) of the Jain. We also noted that, although Amstacandra splits linga into two categories, the external (bahir) and the internal (antara),33 in social terms there can only really be one set of linga, the external, since an individual's internal state can only be demonstrated and evaluated through his external behaviour. With that in mind, it becomes clear that when both Amrtacandra and Jayasena, in their commentaries on this concluding passage from the Samayasāra, say that the linga being referred to are simply dravyalinga, they do little to temper the radical social implications of Kundakunda's total abandonment of linga.34
The first question to be asked of the Samayasāra passage quoted above is who is being referred to by the term pāsamdiya [408 = JGM 438], Jaina or non-Jaina ascetics? According to Monier-Williams, pāşanda is a
32 See pp. 217-222, above. 33 See p. 219.
34 Jayasena makes the distinction between bhāva and dravya linga, i.e. what belongs to the self and what does not, and perhaps also with the sense of internal' and 'external' - see Tātparyavrttih on JGM 438 and 439.

Page #311

Kundakunda: The Samayasāra 297 'heretic'; that is to say, for Hindus it is 'one who falsely assumes the characteristics of an orthodox Hindu' (i.e. a Jaina, a Buddhist, etc.). On first consideration therefore, this would seem to be a term of denigration referring to non-Jaina ascetics (or even to non-Jains in general). Gāthā 411 (= JGM 441), however, uses anagara (literally, 'homeless', and so mendicant or ascetic) as a synonym for pāsamdiya, and anagāra is a standard term for a Jaina ascetic. This may be an example of older verses taking on new meanings in a new context, for gāthā 409 (439) makes it clear that all linga are to be abandoned. In other words, the Samayasāra is not simply making a distinction here between Jaina ascetics and everybody else (perhaps the original meaning of 408 [438]), but between everybody else (lay people and ascetics) and those who have attained kevala-jñāna, the Arhats (see 409 [439], above). This is supported by the Ātmakhyāti on 414 (444) which refers to two kinds of dravyalinga, divided between Jaina ascetics and their lay followers, 35 both of which are vyavahāra views and to be abandoned in favour of the linga-less niścaya view which is the ultimate truth (paramārtha) (see 414 [444] and Ātmakhyāti).
Crucially, the purpose of this passage is not primarily descriptive but injunctive: those addressed are likewise enjoined to give up all linga (i.e. attachment to linga) and thus attain liberation. But who are those addressed? Clearly, they are advanced ascetics who are being urged on by Kundakunda to arhatship, kevala-jñāna and liberation. They are linked by gāthā 414 (=JGM 444) with the niscaya view, i.e. the view that the self in its svabhāva cannot be touched by matter and therefore cannot be bound. So even attachment to ascetic conduct is classified as the product of a vyavahāra view and does not lead one to the highest religious goal. Only self-knowledge - self-realisation - can do that.
35 śramanaśramanopāsakabhedena dvividham dravyalingam - Ātmakhyāti on Samayasāra 414 (444).

Page #312

298 Harmless Souls
To expand on this: the essential self (samayasāra) cannot be realised by those who identify with any linga whatsoever (413 = JGM 443), but it is realised by those who devote themselves to darśana, jñāna and cāritra, which constitute the path to liberation (409 = JGM 411). The vocabulary here is that of the orthodox definition of what comprises the mokṣamarga, given by Umāsvāti at the beginning of his Tattvārtha Sūtra .36 The meaning though carries a substantially different weight; for Kundakunda, as we have seen, has redefined the ratnatraya in terms of selfknowledge: in reality there are no such things as faith (insight), knowledge and conduct, there is just the pure knower. 37 So when he enjoins devotion to faith, knowledge and conduct as the means to liberation (409, etc.), it is understood that this means self-devotion, i.e. concentration and meditation on the self (412). As the Atmakhyāti on Sam. 410 (= JGM 440) puts it:
(Attachment to) dravyalinga is not the path to liberation because, in being dependent on the (material) body it is (attachment to) other substance (paradravya). Therefore, it is faith, knowledge and conduct which are the path to liberation, because in being 38 dependent upon the atman they are one's own substance.
What appears to be a punning allusion to vihāra in gāthā 412 (=JGM 442) makes the internalisation clear: it is the self which is the true vihāra, the real monastery, temple and sacrificial enclosure. Prakrit vihara could be either vihāra or vihara (vi hr) in Sanskrit. The allusion may be intentional; Monier-Williams gives vihāra as a word for a Jain monastery or temple, but the term does not seem to have been widely used by the Jains themselves at this
36
samyagdarśanajñānacāritrāṇi mokṣamārgaḥ - TS 1:1
38
na
37 Samayasāra 16 (=18) - see pp. 234-235, above; cf. pp. 279ff. khalu dravyalingam mokṣamārgaḥ śarīrāśritatve sati paradravyatvāt | tasmād darśanajñānacāritrāṇy eva mokṣamārgaḥ ātmāśritatve sati svadravyatvāt - Ātmakhyāti on JGM 440.

Page #313

Kundakunda: The Samayasāra 299 period, so perhaps the main target here is the Buddhists. The meaning, however, is clear: it is the self alone which constitutes the path to liberation; attachment to external conduct is attachment to what is alien to the self and soteriologically counter-productive.
The essential question now arises: does this mean that the seeker after kevala-jñāna should actually abandon the forms of Jaina ascetic life to concentrate undistractedly on the self? That would be radical indeed, going well beyond the statement of Samayasāra 152-153 (quoted pp. 279-280, above) that tapas, etc., are useless without self-realisation, and taking Kundakunda's perception that ultimately only self-realisation or meditation on the self is soteriologically effective to a logical if not necessary conclusion. The key, however, seems to be that it is attachment to linga that is to be abandoned, not linga itself.39 It is reification that is to be avoided, and the promotion of form over underlying significance. In other words, the ascetic can obtain freedom from linga through internal discipline, the cultivation of an attitude, without having to make external changes. Thus, although internalisation or concentration on the self makes external practice largely irrelevant, the fact that its full logic is not developed until the ascetic is close to kevala-jñāna preserves the formal structure of monastic life.
This emphasis on attachment to the thing rather than the thing itself helps us to understand what, at first sight, appear to be the two most radical, not to say antinomian, gāthās in the Samayasāra, namely, 306 and 307 (= JGM 334 and 335):
Repentance, pursuit of the good, rejection of evil etc., concentration, non-attachment to external objects, self-censure, confession of faults, purification (by expiation) - this is the eightfold pot of poison. (306) Non-repentance, non-pursuit of the good, non-rejection of evil etc.,
39 See Samayasāra 413 = JGM 443.

Page #314

300 Harmless Souls
non-concentration, non-abstinence from attachment to external objects, non-self-censure, non-confession of faults, nonpurification - this is the pot of nectar. (307)
Clearly this eightfold list (306) constitutes a description of the conduct expected of the good ascetic. Although in this form it does not correspond exactly to any formula that I can discover, each of its elements must have been part of monastic practice from an early date. 40
What then can be meant by, for instance, describing the practice of pratisarana (glossed by Jayasena as 'the turning towards qualities such as 'right belief)41 as 'the pot of poison' and its non-performance as 'the pot of nectar'?
The modern commentator, J.L. Jaini,42 explains that, although all eight practices are commendable for ascetics,
for one who is bent solely on the realisation of the self, they are hindrances, and therefore like poison, because they produce bondage of good karmas which keep the soul in Samsara, and stand in the way of its self-realisation.43
He then goes on to state that the practice of self-absorption
40 In my translation of this list I have followed Jayasena's Tātparyavrttiḥ on JGM 334-335. On pratikramana as 'ritualized confession', see JPP p. 349; see also Schubring para. 159. According to the TS (9:22) pratikramana comes under prāyaścitta (repentance), which itself is one of the subdivisions of internal tapas. Kundakunda's Niyamasāra states that meditation (jhāna / dhyāna) on the self constitutes the repentance of all transgressions (pratikramana). See also Niyamasāra 83-91. On parihära, see Schubring para. 161, and TS 9:22, where it occurs in the technical sense of expulsion from the order; cf. Bhargava pp. 185, 189-190. On dhāraņā, see Schubring para. 72; it is also given at TS 1:15 as one of the four divisions of sensory knowledge. On nindā, see Schubring para. 160. On pratikramana, ninda and garhā, see Bhargava p.169.
. 41 samyaktvādiguneșu preranam - Tātparyavrtti on Samayasāra 306 (= JGM 334].
42 On Samayasāra 306 (= JGM 334, and 327 in Jaini's SBJ ed.) 43 p. 174, SBJ ed.

Page #315

Kundakunda: The Samayasāra 301
is 'to remove the perverse belief that practical conduct of saints, i.e. repentance, etc., will lead to Liberation from bondage of karmas'.44 In other words, these gāthās describe an attitude to be adopted by those on the verge of pure knowledge and liberation; and they are articulated in a way which is probably designed to shock ascetics out of attachment to those elements of their religious life which they hold most dear. (In this respect these gāthās are perhaps comparable in method to the kōans of Ch'an Buddhism.) Moreover, the eight practices given at Samayasara 306 are connected with lapses from perfect monastic conduct: the need for them implies lack of perfection on the part of the ascetic. Similarly, when they are not needed it is because perfection has been attained. (And once again ontology and epistemology do their dance: the state of perfection, where prescribed ascetic practices are transcended, is finally achieved by non-attachment to - i.e. by transcendence of - precisely such practices.)
Another modern commentator, Chakravarti, takes these gāthās to be descriptive (i.e. broadly philosophical) rather than prescriptive (religious) in tenor; he sees them as representing the perspective of the already purified 'transcendental Self, beyond good and evil, for whom the question of discipline or non-discipline is quite meaningless. For Chakravarti, the term apratikramaṇa does not imply the mere opposite of pratikramana (which would imply the removal of discipline and giving full reign to the impure emotions), rather the negative prefix (a-) 'must be taken to signify the absence of necessity to practise the discipline'.45 For the self absorbed in its own pure nature impure psychic states are brought to a stop, so it is unnecessary to practise the various kinds of discipline.46
This interpretation saves him from the potentially
44 Ibid.
45
p.190, Chakravarti's ed. of Samayasāra. 46 See ibid. pp. 189-190.

Page #316

302 Harmless Souls embarrassing concession that it is possible, even necessary, to give up orthodox ascetic practices short of actual liberation. The purpose of the Samayasāra, however, is religious; and both here (Samayasāra 306 and 307) and in the concluding gāthās (408-415) the intention is clearly to engender an attitude of total non-attachment in the listener. Indeed, in the very last verse (415) it is stressed that those who understand and hold to the true meaning will, for those very reasons, attain the highest bliss. This implied devaluation of actual ascetic practice - from the highest perspective (the second niscaya view) it too is a hindrance to liberation - obviously has the potential to threaten that conduct which is the defining basis of Jajna religion. (In soteriological matters the inessential quickly becomes the irrelevant: what is no longer essential for liberation becomes a hindrance to be abandoned.)
It is little wonder, therefore, that the Samayasāra was at one time considered a text 'too sacred to be read by householders'.47 Upadhye states that 'the spiritual statements from Niscaya-naya may prove socially and ethically harmful to the house-holders who are almost absolutely lacking in spiritual discipline!.48 But it is also clear that, unless tempered by other instruction, such teachings must have also posed a threat to ascetic behaviour. The full extent of that threat perhaps only becomes apparent in such Apabhramsa works as Yogindu's Paramātmaprakāśa (c. 900 C.E.?), and research is clearly required in that field before the limits of Jaina orthodoxy can be accurately defined.
In the works attributed to Kundakunda, and in the Samayasāra in particular, a point of maximum tension is reached between soteriological theory and social necessity, between the inner state and the outer discipline, between ātman and linga. But by the time of Yogindu, the connection seems to have snapped for some, and orthodoxy
47 Upadhye p. xlvii. 48 Ibid.

Page #317

Kundakunda: The Samayasara 303
(i.e. external religious practice) has been largely abandoned; for others, ascetic practice retains its dominance as the emblem and preserver of Jaina identity. The significance of Kundakunda's works is that at one level they point in the direction that was to be taken by Yogindu while at another they prescribe largely conventional practices. Their eclectic nature entails internal contradictions, but it is these very contradictions which give commentators scope to temper the full (heretical) implications of the niscaya strand in Kundakunda's thought. Thus it is ensured that, in practice, the threat to Jaina conduct and social identity is circumscribed and remains latent rather than actual.
9.3 Liberation according to Kundakunda: some conclusions
As we have seen, in Kundakunda's soteriological theory liberation is a matter of knowledge (jñāna), of realisation of the true nature of the self through meditation. But in Digambara as in Śvetāmbara practice it remains a matter of largely external asceticism, the maintenance of (physical) ahimsā. This is because it is external practice that distinguishes 'Jainism' as a religious tradition. Whatever the philosophical rationale for doing so, such practices cannot be abandoned without threatening the Jaina's sense of identity. Jainism's life crisis rituals, which strictly speaking are irrelevant to soteriology, are largely taken from Hindu models;49 its doctrines, if taken to their logical conclusions (as they sometimes are by Kundakunda), become so internalised and karma is so dematerialised that the dividing line between such doctrines and those, for instance, of Sāmkhya and Vedanta becomes attenuated to the point of non-existence (except for a few, logically arbitrary points). It becomes crucial, therefore, to retain at some level the reality of the connection between the soul
49 See JPP pp. 291-304; R. Williams 1963, pp. 274-287; Sangave pp. 243-252, 381.

Page #318

304
Harmless Souls
and physical karman, and so keep tapas as the primary means to liberation, otherwise the link with tradition is broken and the monk's discipline pointless. It is the logicians or scholastics, obliged to express themselves in broadly philosophical terms, who are confronted by the full force of this problem.50
Jaina doctrines and ascetic practices were originally formulated in the context of physicalist or materialist ideas, probably including the notion of a material or quasimaterial soul. Later, the soul is conceived of as fundamentally immaterial, but the means of bondage (karman) remains unambiguously material. The logical tensions which arise from this juxtaposition of the material and the non- material are stripped bare at times of dispute and philosophical 'system-building'. The scholastics have to retain the tradition, embodied in the practices of the ascetic, and at the same time provide a justification or doctrinal rationale for such behaviour. They also have to deal with the accumulated and accumulating rationalisation and internalisation of doctrine, derived ultimately from the contradiction between the material and immaterial at the heart of that teaching. Such a contradiction invites extreme solutions. But given prevailing Indian beliefs about the nature of the soul, and the inherent tendency of doctrines to become more rather than less sophisticated to move from ethics to metaphysics a return to the idea of a material soul was unlikely. In addition, there were other, perhaps more important reasons why the movement should have been towards greater and greater internalisation, since, at
-
50 By logicians or scholastics I mean fifth century (?) writers such as Siddhasena and Samantabhadra (see JPP p. 83ff.) and those who followed them. For a list of such writers and summaries of their main works through to the seventeenth century, see Dixit 1971, pp. 88-164. Dixit characterises this period as the 'Age of Logic'. Undoubtedly, some of the works attributed to Kundakunda were at least compiled in this period (Dixit includes him in his list), but in terms of content they are far less orthodox, as Dixit himself points out when he refers to the Samayasāra (see pp. 93-94, 132-135).

Page #319

Kundakunda: The Samayasāra 305 least on the surface, this strategy provides the 'easier' or graded route to liberation, and, as we have seen, lowers the soteriological ladder into the territory of the laity. However, as has been made clear, to take this process to its logical conclusion threatens to disintegrate the tradition, and with it 'Jaina' identity. The scholastics, therefore, have to maintain the tension without falling into either camp (the real danger, of course, being from the kind of total internalisation, with its attendant 'dematerialisation of karman, which Kundakunda approaches in parts of the Samayasāra). Given this, I would suggest that a major reason for the peculiar content and form of what is technically considered to be 'Jaina philosophy' - syādvāda, anekāntavāda, and the (orthodox) naya doctrine - is precisely the need to retain a raison d' être for the ascetic practices which constitute Jaina identity in the face of a progressive tendency to rationalise and internalise doctrinal formulations.
In the works ascribed to Kundakunda, however, internalisation threatens at times to break the controls imposed by Jaina philosophy. If the 'orthodox' naya doctrine allows neo-Vedāntic teachings to be propagated (at one level) without precipitating a collapse into the vacuum created by the implications of a non-material soul doctrine, Kundakunda's heterodox reading of the niscaya-vyavahāra distinction in the Samayasāra) removes that philosophic restraint. By doing so it provides a glimpse of the full, 'heretical' implications for Jaina practice and doctrine of radical internalisation deriving from the conviction that the soul really has no connection with matter whatsoever. For the predominant view in Kundakunda is that what counts soteriologically is what happens internally, in consciousness. Instrumental in this are upayoga (in the Pravacanasāra) and bhāva (in the Samayasāra); stress thus falls on meditation (dhyāna) leading to realisation (jñāna) of the self's total separation from matter - indeed, the pure self is defined in both the Pravacanasāra and the Samayasāra as (such) knowledge - rather than on the

Page #320

306 Harmless Souls practice of tapas. Consequently, cāritra comes to be seen in terms of 'attitude' rather than physical action / inaction.
For ascetics, himsā is redefined as harm caused (internally) to the self through neglect of or offence against the monastic rules. This is significant for two paradoxical, although ultimately complementary reasons. On the one hand, it permits the 'formalisation' or 'ritualisation of the monks' conduct: that is to say, it is now fulfilment of the role, not behaviour as such, that counts, for by following the prescribed rules the desired end will inevitably be achieved. On the other hand, it opens the way for intention to become the the chief instrument of bondage and liberation. To put it differently, himsā is only himsā if it is intended; but, by definition, as long as one's behaviour accords with the rules laid down for ascetics then himsā cannot be intended. Therefore, as the texts remark, there can be no himsā for proper ascetics regardless of the harm done to living creatures. When this formulation is combined with Kundakunda's upayoga doctrine - and to some extent the former clears the way for the latter - then, from that internalised perspective, it becomes clear that it is impure consciousness which leads to wrong intention and thus ultimately to violation of the ascetic rules. There can be no violation by accident; transgression has to be preceded by intention, and thus by impure consciousness. So aśuddhopayoga is at the root of all himsā, and all himsā is himsā to one's self.
Seen from this perspective, external, physical behaviour in the world is essentially irrelevant to soteriology: bondage and liberation begin and end with consciousness, external conduct is merely the outward sign of an inner state. Thus on the one hand there is the belief that simply by following the external rules to the letter one will achieve the religious goal, and on the other there is the belief that the only soteriologically significant behaviour is that which brings about inner purity. But by asserting that if the monk is following his rules exactly then ipso facto he must be internally pure, one brings about a meeting or balance

Page #321

Kundakunda: The Samayasāra 307
between the extremes of total internalisation and total externalisation; one becomes a reflex of the other, and depending upon which component is given causal preference or emphasised, the external or the internal, a different view is obtained of the institution of renunciation.
We may call these views the 'social' and the 'personal'; the former is to do with questions of Jaina identity (for the Jains themselves, and therefore necessarily vis à vis the wider community), the latter is a question of personal soteriology for individuals (śramaṇas and advanced laity). In Kundakunda and his commentators, when the question is raised of why one should go on with physical asceticism when everything of soteriological significance is internal, a matter of realisation, the reply is that one should continue to behave in the prescribed ways because they provide an outward sign of inner purity. Philosophically this is less than compelling, but it is a line of argument with an underlying social imperative; for, as we have seen, what distinguishes Jainism as a religious tradition is the behaviour of its ascetics. So to behave in such a way is not merely the external sign of an ideal Jaina religious, it is actually to embody and perpetuate the unique socioreligious identity of the Jaina community. From the point of view of the laity, therefore, the function of the ascetic is to provide them with a symbol or emblem of their own tradition, one which it is essential to preserve. And this is probably the underlying motivation for the laity's diligence in ensuring that the śramanas perform their ascetic practices to the letter.51 (The ascetic also, of course, provides the laity with a constant reminder of the soteriological ideal towards which they may more or less urgently aspire.)
This is the 'social' view of renunciation which I have connected with external ascetic behaviour. The 'personal' view I see as being concerned with internal transformation and gnosis, as exemplified in the teachings of Kundakunda,
51 See JPP

P. 208.

Page #322

308 Harmless Souls i.e. with soteriology. These two views do not necessarily coincide with the further distinction between the views of the laity and those of the śramaņas themselves. It is clear that both originally and throughout most of Jaina history the soteriological emphasis for ascetics and advanced lay followers falls upon external practice. Thus internal practices may be taken on as extra vrata, or because the correct internal attitude is required before external practice is effective, but they do not replace external tapas. One may surmise that for the majority of ascetics the practical implication of the internalisation of doctrine - viz. liberation through self-realisation - itself remained an ideal, to be achieved or even attempted only by the most advanced śramaņas. Most hoped to approach liberation and attain at least a better rebirth by following the rules for ascetics to the letter (from the 'social' view their proper role), a formidably difficult undertaking even in its ritualised form.
There are indications in the texts that this process of ritualisation started at a relatively early date.52 From one perspective this may be seen as a quest for greater autonomy on the part of the renouncers, the need for their soteriological progress to be in their own hands and less subject to the fortuitous and accidental. The drive towards greater and greater control over the process of one's own liberation is probably also one of the contributory reasons for a progressive internalisation of practice. Furthermore, it may be remarked that through 'ritualising' external behaviour the way is cleared for a greater concentration on obtaining inner purity. For simply by following the rules the monk's external considerations are taken care of; he does not have to 'think about his behaviour, and so his consciousness can be engaged elsewhere, in the realisation
52 E.g. in Viyāhapannatti, Sarvārthasiddhi and in Book 3 of the Pravacanasāra, parts of which are probably derived from older material

Page #323

Kundakunda: The Samayasāra 309 of his own inner nature or self. And it may be that the doctrines expounded by Kundakunda were in part intended as an antidote to excessive 'ritualisation', a reformation through radical internalisation, emphasising that it is not the external form that counts but the internal transformation of consciousness. Such a reading is borne out by the fact that this teaching was aimed specifically at śramaņas, i.e. at those who had to some extent already mastered the external forms.
We can see from the above that, although doctrines such as Kundakunda's undercut external asceticism at a theoretical level to the extent that the rationale for such asceticism becomes slender, there are overriding social reasons which ensure the retention of traditional practice. For it is the practice of ascetics which carries the tradition and which provides the Jainas with their socio-religious identity. With the acknowledged loss of most of their original canon, such ascetic practice is thus a particularly important vehicle for the Digambaras. Consequently, although Kundakunda is acknowledged and revered as being virtually the initiator or compiler of their 'new' scriptural and doctrinal tradition, the implications of his less orthodox teachings are not permitted to destabilise or threaten established ascetic practice; for the tradition that provides the Digambaras with their identity is already in place, embodied in that same ascetic practice, in the behaviour and monastic rules of the śramaņas themselves. So while śuddhopayoga or the pure self may become the inner emblem or ideal of the aspiring ascetic, it does not replace or negate the outer emblem of ascetic practice. The new philosophical and doctrinal reasons for this retention may be weak but the social or communal reasons are stronger.
In addition, it should be remembered that the spiritually advanced laity also undertake numerous ascetic practices, so they too carry the tradition. Thus although internalisation of doctrine gives them a foothold on the soteriological ladder, the actual ascent requires external

Page #324

310 Harmless Souls practice, with the full 'logic' of internalisation coming to the fore only for those to whom external ascetic practice is already second nature. To put it another way, it may be the case that external practice itself represents the path to inner purity for the laity and less advanced ascetics, while for the ideal ascetic it is merely the outer sign of an already achieved inner state. In this fashion the recurrent ambiguity of ritual is demonstrated, for the question of whether it is instrumental or expressive is not answered: it is simply ambiguous.

Page #325

Table
samrambha (impulsion / determination to do violence)
can be done by the 3 yogas speech activity
samārambha (collecting means / preparation
to do violence)
can be done by the 3 yogas
ārambha (actually undertaking
violence / commencement of violence)
bodily activity
can be done by the 3 yogas
mind activity
oneself
can be done by one's agent
can be done by ibid
ETC.
others with one's approval
can be done by ibid
krodha (anger) măna (pride) due to māyā (deceitfulness) lobha (greed) due to ibid
due to ibid
[= 108 ways in which a jiva is counted as the adhikarana of asrava - i.e. 108 ways in which the jiva can cause injury]

Page #326

Appendix 1
Niyamasāra
18
kattā bhottā ādā poggalakammassa hodi vavahāro | kammajabhāveņādā kattā bhottā du ņicchayado ||
103
jam kimci me duccarittam savvam tiviheņa vosare | sāmāiyam tu tiviham karemi savvam ņirāyāram ||
104
sammam me savvabhūdesu veram majjham na keņavi| āsāe vosarittānam samāhi pațivajjae ||
133
jo du dhammam ca sukkam ca jhāņam jhāedi ņiccasā|| tassa sāmāigam thāi idi kevalisāsaņe ||
159 (SBJ 158)
jāņadi passadi savvam vavahāraṇaeņa kevali bhagavam | kevalaņāņī jāņadi passadi niyameņa appāņam ||

Page #327

Appendix 2
Pañcāstikāya
115
sammattam saddahaņam bhāvāṇam tesim adhigamo ņāņam || cārittam samabhāvo visayesu virūdhamaggānam II
147 (SBJ 154)
jam suhamasuham udiņnam bhāvam ratto karedi jadi appā || so teña havadi bamdho poggalakammeņa viviheņa ||
148 (SBJ 155)
joganimittam gahaņam jogo maṇavayaņakāyasambhūdo. bhāvanipitto bamdho bhāvo radiragadosamohajudo ||
162
jivo sahāvaņiyado aņiyadagunapajjao dha parasamao jadi kunadi sagam samayam pabbhassadi kammabamdhādo ||

Page #328

Appendix 3
Pravacanasāra
1:7
cārittam khalu dhammo dhammo jo so samo tti ņiddittho | mohakkhohavihīņo pariņāmo appaņo hu samo ||
1:9
jīvo pariņamadi jadā suheņa asuheņa vā suho asuho || suddheņa tadā suddho havadi hi pariņāmasabbhāvo ||
1:11
dhammeņa pariņadappā appā jadi suddhasampayogajudo | pāvadi ņivvāņasuham suhovajutto va saggasuham ||
1:23
ādā ņāņapamāṇam ņāņam ņeyappamāṇam uddiţtham ņeyam loyāloyam tamhā ņāņam tu savvagayam ||
1:26
savvagado jiņavasaho savve vi ya taggayā jagadi atthā| ņāņamayādo ya jiņo visayādo tassa te bhaniyā |||
1:27
ņāņam appa tti madam vattadi ņāņam viņā ņa appāņam | tamhā ņāņam appā appā ņāņam va annam vā ||
1:29
ņa pavittho nāvisho ņāņi ņeyesu rūvam iva cakkhū|
jāņadi passadi niyadam akkhātido jagam asesam || Upadhye takes na pavittho nāvittho as Sanskrit. na praviştah na āvistaḥ, as against the commentators who take na aviştah (na apravistah) - see his fn. 2, p. 4.

Page #329

Appendix 3 Pravacanasāra 315
1:30
rayaņam iha imdaņīlam duddhajjhasiyam jahā sabhāsāe | abhibhūya tampi duddham vaťţadi taha ņāņam atthesu ||
1:32
genhadi ņeva ņa mumcadi ņa param pariņamadi kevali bhagavam | pecchadi samamtado so jāņadi savvam ņiravasesam ||
1:35
jo jāņadi so ņāņam ņa havadi ņāņeņa jāņago ādā | ņāņam pariņamadi sayam atthā ņāņatthiyā savve ||
1:43
udayagadā kammamsā jiņavaravasahehim niyadiņā bhaniyā || tesu vimūdho ratto duttho vā bamdham anubhavadi ||
1:46
jadi so suho va asuho na havadi ādā sayam sahāveņa | samsāro vi ņa vijjadi savvesim jīvakāyāṇam ||
1:52
ņa vi pariņamadi ņa geņhadi uppajjadi ņeva tesu asthesu | jānannavi te ādā abamdhago teņa paņņatto ||
1:69
devadajadigurupājāsu ceva dāņammi vā susilesu | uvayāsādisu ratto suhovaogappago appā ||
1:72
naraņārayatiriyasurā bhajamti jadi dehasambhavam dukkham | kiha so suho va asuho uvaogo havadi jīvānām ||
1:77
na hi mannadi jo evam ņatthi viseso tti puņnapāvāņam|| himdadi ghoram apāram samsāram mohasamchaạno ||

Page #330

316 Appendix 3 Pravacanasāra
1:78
1:85
2:29
2:30
2:58
2:59
2:63
2:64
evam vididattho jo davvesu na rāgam edi dosam vā uvaogavisuddho so khavedi dehubbhavam dukkham ||
2:65
atthe ajadhāgahaṇam karuṇābhāvo ya maṇuvatiriesu | visaesu a ppasamgo mohassedāṇi limgāṇi ||
ādā kammamalimaso pariņāmam lahadi kammasamjuttam tatto silisadi kammam tamhā kammam tu pariņāmo ||
pariņāmo sayam ādā sā puņa kiriya tti hodi jīvamayā | kiriyā kamma tti madā tamhā kammassa na du kattā ||
ādā kammamalimaso dharedi pāṇe puno puno anne ṇa cayadi jāva mamattam dehapadhānesu visayesu ||
jo imdiyādi vijai bhavīya uvaogam appagam jhādi | kammehim so na ramjadi kiha tam pāṇā aṇucaramti ||
appā uvaogappā uvaogo ṇāṇadamsanam bhanido | so vi suho asuho va uvaogo appano havadi ||
uvaogo jadi hi suho punnam jivassa samcayam jādi | asuho va tadhā pāvam tesim abhāve na cayam atthi || tadha (line 2) emended to tadhā
jo jāṇādi jiņimde pecchadi siddhe taheva anagare | jivesu sāņukampo uvaogo so suho tassa ||

Page #331

Appendix 3 Pravacanasāra 317
2:67
asuhovaogarahido suhovajutto na anņadaviyamhi hojjam majjhattho 'ham ņāņappagam appagam jhāe ||
2:68
ņāham deho ņa maño ņa ceva vāņi ņa kāraṇam tesimi kattā ņa ņa kārayidā aṇumamtā ņeva kattīņam ||
2:80
arasam arūvam agamdham avvattam cedaņāgunam asaddam | jāņa alimgaggahaņam jīvam aạidditthasamthāņam |||
2:81
mutto rūvādiguņo bajjhadi phāsehim annamaņņehim tavvivarido appā bajjhadi kidha poggalam kammam ||
2:83
uvaogamao jīvo mujjhadi rajjedi vā padussedi | pappā vividhe visaye jo hi puņo tehim sambamdho ||
2:86
sapadeso so appā tesu padesesu puggalā kāyā|| pavisamti jahājoggam citthamti hi jamti bajjhasti ||
2:87
ratto bamdhadi kammam muccadi kammehim rāgarahidappā|| eso bamdhasamāso jīvāņām jāņa Ņicchayado ||
2:88
pariņāmādo baņdho pariņāmo rāgadosamohajudo asuho mohapadoso suho va asuho havadi rāgo ||
2:89
suhapariņāmo punnam asuho pāva tti bhaniyam aạnesu | pariņāmo nannagado dukkhakkhayakāraṇam samaye ||

Page #332

318 Appendix 3 Pravacanasāra 2:90
bhaạidā pudhavippamuhā jīvaņikāyādha hāvarā ya tasā|| aņņā te jīvādo jīvo vi ya tehimdo aạno ||
2:91
jo ņavi jāņadi evam paramappāņam sahāvam āsejja kiradi ajjhavasāņam aham mamedam ti mohādo ||
2:92
kuvvam sabhāvam ādā havadi hi kattā sagassa bhāvassa | poggaladavvamayānam ņa du kattā savvabhāvāņam |||
2:93
genhadi ņeva na mumcadi karedi na hi poggalāņi kammāņi | jivo puggalamajjhe vattaņņ avi savvakālesu |||
2:97
eso bamdhasamāso jīvāņam nicchayeņa şiddittho 1 arahamtehim jadīņam vavahāro annahā bhaņido ||
2:98
ņa cayadi jo du mamattim aham mamedam ti dehadaviņesu | so sāmaņņam cattā paļivaņņo hodi ummaggam ||
2:99
ņāham homi paresim ņa me pare santi ņāņam aham ekko idi jo jhāyadi jhāṇe so appāņam havadi jhādā ||
2:101
dehā vā daviņā vā suhadukkhā vādha sattumittajaņā| jīvassa ņa samti dhuvā dhuvovaogappago appā ||
2:102
jo evam jāņittā jhādi param appagam visuddhappā || sāgāro 'ņāgāro khavedi so mohaduggamthim ||

Page #333

Appendix 3 Pravacanasāra 319
2:104
jo khavidamohakaluso visayaviratto maño ņirumbhittā | samavasthido sahāve so appāņam havadi jhādā ||
2:108
tamhā taha jāņittā appāņam jāņagam sabhāveņa | parivajjāmi mamattim uvatthido nimmamattammi ||
3:5
jadhajādarūvajādam uppādidakesamamsugam suddham | rahidam himsādīdo appaļikammam havadi limgam ||
3:6
mucchārambhavimukkam juttam uvajogajogasuddhihim limgam ņa parāvekkham apunabbhavakāraṇam jeņham ||
3:16
apayattā vā cariyā sayaņāsaņathāṇacamkamādīsu | samaņassa savvakāle himsā sā samtattiya tti madā ||
3:17
maradu va jiyadu va jīvo ayadācārassa nicchidā himsā |
payadassa ņatthi bamdho himsāmetteņa samidassa il All editions have jiyadu, although the chāyā gives jīvatu.
3:17b
uccāliyamhi pāe iriyāsamidassa niggamatthāe | ābādhejja kulimgam marijja tam jogam āsejja ||
3:17c
na hi tassa tannimitto bamdho suhumo ya desido samaye | mucchā pariggaho cciya ajjhappapamāṇado dițțho ljummamil
3:18
ayadācāro samaño chassu vi kāyesu vadhakaro tti mado caradi jadam jadi ņiccam kamalam va jale ņiruvalevo ||

Page #334

Appendix 3 Pravacanasāra
320 3:19
havadi va na havadi bamdho madamhi jive 'dha kāyacetthamhi bamdho dhuvam uvadhido idi samaņā chaddiyā savvam ||
3:20
ņa hi ņiravekkho cāgo na havadi bhikkhussa āsayavisuddhi| avisuddhassa ya citte kaham ņu kammakkhao vihio ||
3:20c
vatthakkhamdam duddiyabhāyaṇam annam ca geņhadi niyadam | vijjadi pāņārambho vikkhevo tassa cittammi:||
3:27
jassa anesaņam appā tam pi tayo tappadicchagā samaņā | annam bhikkham anesaņam adha te samaņā aṇāhārā ||
3:45
samaņā suddhuvajuttā suhovajuttā ya homti samayamhi | tesu vi suddhuvajuttā aṇāsavā sāsavā sesā ||
3:54
esā pasatthabhūdā samaņāņam vā puno gharatthāņam | cariyā paretti bhaņidā tāeva param lahadi sokkham ||
3:74
suddhassa ya sāmaņņam bhaniyam suddhassa damsaņam ņāņam | suddhassa ya nivvānam so cciya siddho namo tassa ||

Page #335

Appendix 4
Samayasāra Text compiled from Chakravarti's and JGM editions, following Chakravarti's numbering system unless otherwise stated. The bracketed numerals refer to the Sanātana Jaina Grantha Mala (JGM) and, in one instance, the Sacred Books of the Jainas (SBJ) equivalents.
2 (2)
jivo carittadamsaņaņāņatthido tam hi sasamayam jāņa poggalakammuvadesatthidam ca tam jāņa parasamayam ||
3 (3)
eyattaņicchayagado samao savvattha sundaro loge 1 bamdhakahā eyatte teņa visamvādiņi hodi ||
vavahāreņuvadissadi ņāņissa carittadamsaņam ņāņam | navi ņāņam ņa carittam na damsaņam jānago suddho ||
8 (8)
jaha ņavi sakkam anajjo anajjabhāsam viņā u gāhedum | taha vavahāreņa viņā paramatthuvadesaņam asakkam ||
11 (13)
vavahāro 'bhūdattho bhūdattho desido du suddhaņao | bhūdattham assido khalu samādithi havadi jivo ||
11 (JGM only)
ņāņamhi bhāvaņā khalu kādavvā damsaņe caritte ya| te puņa tiņņi vi ādā tamhä kuņa bhāvanam āde 11
12 (14)
suddho suddhādeso ņādavvo paramabhāvadarisīhim| vavahāradesido puņa je du aparame thidā bhāve ||

Page #336

322 Appendix 4 Samayasāra
12 (JGM only)
jo ādabhāvaņam iņam ņiccuvajutto muņī samācaradi | so savvadukkhamokkham pāvadi acireņa kālena ||
14 (16)
jo passadi appāņam abaddhapuţtham anannayam niyadam| avisesam asamjuttam tam suddhaņayam viyāņihi ||
16 (19)
damsaņaņāņacarittāņi sevidavvāņi sāhuņā ņiccam| tāņi puna jāņa tiņņi vi appāņam ceva nicchayado ||
19 (22)
kamme nokammamhi ya aham idi ahayam ca kamma ņokammam jā esā khalu buddhi appadibuddho havadi tāva ||
24 (JGM only)
jam kuņadi bhāvam ādā kattā so hodi tassa bhāvassa || ņicchayado vavahārā poggalakammāņa kattāram ||
32 (37)
jo moham tu jiņittā ņāṇasahāvādhiyam* munadi ādam |
tam jidamoham sāhum paramasthaviyāņayā vimti || *All editions give adhiyam, which looks doubtful.
34 (39)
ņāņam savve bhāve paccakkhādi ya pare tti ņādūņa | tamhā paccakkhāņam ņāņam niyamā muñedavvam ||
36 (41)
ņatthi mama ko vi moho bujjhadi uvaoga eva aham ekko tam mohanimmamattam samayassa viyāņayā viņti ||
38 (43)
aham ekko khalu suddho damsaņaņāṇamaio sadārūvi | navi atthi majjha kimcivi annam paramāņumittam pi ||

www.jaintelibrary.org

Page #337

Appendix 4 Samayasāra
323
46 (51)
vavahārassa darisanam uvadeso vaņņido jiņavarehim |
jīvā* ede savve ajjhavasāņādao bhāvā ||| * Although the texts print jīvā, nominative, this makes no real sense and should, I suggest, be emended to jīve.
71 (76)
jaiyā imeņa jiveņa appaņo āsavāņa ya taheval. ņādam hodi visesamtaram tu taiyā ņa bamdho se ||
72 (77)
ņādūņa āsavāṇam asucittam ca vivariyabhāvam ca dukkhassa kāraṇam ti ya tado niyattim kuņadi jīvo ||
83 (89)
ņicchayaņayassa evam ādā appāņam eva hi karedi | vedayadi puño tam ceva jāņa attā du attāṇam| -
84 (90)
vavahārassa du ādā poggalakammam karedi aņeyaviham | tam ceva ya vedayade poggalakammam aneyaviham ||
85 (91)
jadi poggalakammam iņam kuvvadi tam ceva vedayadi ādā | dokiriyāvādittam pasajadi so jiņāvamadam ||
86 (92)
jamhā du attabhāvam poggalabhāvam ca do vi kuvvati | teņa du micchādiţthi dokiriyāvādiņo honti ||
92 (99)
param appāņam kuvvadi appāņam pi ya param karamto so aņņāṇamao jīvo kammāņam kārago hodi ||
93 (100)
param appāņam akuvvi appāņam pi ya param akuvvamto so ņāṇamao jīvo kammāņam akārago hodi ||

Page #338

324
Appendix 4 Samayasāra
127 (137)
aņņāņamao bhāvo aņāņiņo kuņadi tena kammāņi| ņāņamao ņāņissa du ņa kunadi tamhā du kammāņi ||
141 (151)
jīve kammam baddham puttham cedi vavahāraṇayabhanidam | suddhaņayassa du jive abaddhaputtham havai kammam ||
142 (152)
kammam baddham abaddham jīve edam tu jāņa ņayapakkham pakkhātikkamto puņa bhannadi jo so samayasāro ||
143 (153)
doņha vi ņayāņa bhaạidam jāņai ņavarim tu samayapaļibaddho |
ņa du ņayapakkham giņhadi kimci vi ņayapakkhaparihiņo* || *Sheth gives parihīna.
144 (154)
sammaddamsaņaņāņam eso (= edam JGM] lahadi tti navari vavadesam |
savvaņayapakkharahido bhaạido jo so samayasāro || 151 (161)
paramațțho khalu samao suddho jo kevali muņi ņāņi | tamhi thidā sahāve muņiņo pāvamti ņivvāņam
152 (162)
paramasthammi ya athido jo kunadi tavam vadam ca dhārayadi
tam savvam bālatavam bālavadam viņti savvahņu* || Savvahņu looks odd. Pischel, para. 105, gives savvannu for sarvajñā. He also gives an instance of savvanhu, which should perhaps be the form here.

Page #339

Appendix 4 Samayasāra
325
153 (163)
vadasiyamāņi dharamtā silāņi tahā tavam ca kuvvamtā | paramațțhabāhirā jeņa teņa te homti aņņāņi ||
156 (166)
mottūņa ņiccayattham vavahāre na vidusā pavatthamti |
paramatham assidāņam du jadīņa kammakkhao vihio || JGM 166 prints hodi for vihio. I have printed pavatthamti as this is the version given in all editions and commentaries, although there seems to be no linguistic explanation for the presence of the aspirate.
183 (176)
edam tu avivarīdam ņāņam jaiyā du hodi jivassa taiyā ņa kimci kuvvadi bhāvam uvaogasuddhappā ||
185 (178)
evam jāņadi ņāņi aņņāņi muñadi rāgam evādam| aņņāņatamocchaņņo ādasahāvam ayāṇamto ||.
186 (179)
suddham tu viyāṇamto suddham evappayam lahadi jivo! jāņamto du asuddham asuddham evappayam lahadi ||
187 (180)
appāņam appaņo rumbhidūņa dosu pun apāvajogesu damsaņaņāņamhi tthido icchāvirado ya asnamhi ||
188 (181)
jo savvasamgamukko jhāyadi appāņam appaņo appā || ņavi kammam nokammam cedā cistedi eyattam ||
189 (182)
appāņam jhāyamto damsaņaņāṇamao añaņņamao || lahadi acireņa appāņam eva so kammaņimmukkam ||

Page #340

326
Appendix 4 Samayasāra
198 (213)
udayavivāgo viviho kammāņam vannido jiņavarehim | ņa du te majjha sahāvā jāņagabhāvo du aham ekko ||
200 (212)
evam sammāitthi appāņam munadi jāņagasahāvam | udayam kammavivāgam ca muadi taccam viyāņamto ||
201 (214)
paramāņumittiyam pi hu rāgādiņam tu vijjade jassa | ņavi so jāņadi appāņayam tu savvāgamadharo vi ||
202 (-)
appāņam ayāņamto anappayam ceva so ayāņamto kaha hodi sammadithi jīvājīve ayāṇamto |
203 (219)
ādamhi davvabhāve athire mottūņa ginha tava niyadam | thiram ekam imam bhāvam uvalabbhamtam sahāveņa ||
204 (223)
ābhiņisudohimaņakevalam ca tam hodi ekkam eva padam | so eso paramattho jam lahidum ņivvudim jādi ||
205 (224)
ņāņaguņehim vihīņā edam tu padam bahū vi ņa lahamti|| tam ginha supadam edam jadi icchasi kammaparimokkham ||
210 (225)
apariggaho añiccho bhanido ņāņi ya ņicchade dhammam apariggaho du dhammassa jāņago teņa so hodi ||
211 (226)
as 210, except adhamma is substituted for dhamma
212 (228)
as 210, except asaņa is substituted for dhamma

Page #341

213 (229)
as 210, except pāņa is substituted for dhamma
214 (230)
evamādu* edu vivihe savve bhāve ya nicchade ṇāṇi | jāṇagabhāvo niyado niralambo du** savvattha || Jayasena has* ivvādu and **
ya
247 (265)
jo mannadi himsāmi ya himsijjāmi ya parehim sattehim | so mūḍho anṇāņi nāṇī etto du vivarido ||
262 (280)
ajjhavasideņa bandho satte mareu mā va māreu | eso bandhasamāso jīvānam nicchayaṇayassa ||
263 (281)
Appendix 4 Samayasara 327
evam aliye adatte abramhacere pariggahe ceva | kirai ajjhavasāṇam jam tena du bajjhae pāvam ||
264 (282)
tahavi ya sacce datte bambhe aparigahattāne ceva | kirai ajjhavasāṇam jam tena du bajjhae punṇam ||
265 (283)
vatthum paducca jam puna ajjhavasāṇam tu hodi jīvāṇam | na ya vatthudo du bamdho ajjhavasänena bamdho tti||
270 (294)
edāņi natthi jesim ajjhavasāṇāņi evam ādiņi |
te asuhena suheņa va kammeņa muni na lippanti ||
271 (295)
buddhi vavasão vi ya ajjhavasāṇam madi ya viņṇāṇam | eyattam eva savvam cittam bhavo ya parināmo ||

Page #342

328 Appendix 4 Samayasāra
272 (296)
evam vavahāraṇao paạisiddho jāņa Ņicchayaņayeņa |
nicchayaņayassidā puņa muniņo* pāvanti ņivvāņam || *JGM has the alternative reading: nicchayanayasallīņā muniņo
293 (321)
bandhānam ca sahāvam viyāņio* appaņo sahāvam ca
bamdhesu jo virajjadi** so kammavimokkanam kuņai || JGM has *viyāṇidum and **jo na rajjadi
294 (322)
jivo bamdho ya tahāchijjamti salakkhaņehim niyaehim | paņņāchedanaeņa u chiņnā ņāņattam āvannā ||
295 (323)
jīvo bamdho ya tahā chijjamti salakkhanehim niyaehim | bamdho cheyayavvo suddho appā ya ghittavvo 11
296 (324)
kaha so ghippai appā pannāe so u ghippae appā | jaha pannāe yibhatto taha pannā eva ghittavvo ||
297 (325)
paņņāe ghittavvo jo cedā so aham tu ņicchayado | avasesā je bhāvā te majjha pare tti ņāyavvā ||
298 (326)
pannāe ghittavvo jo datthā so aham ... ibid. ||
299 (327)
paņņāe ghittavvo jo ņādā so aham ... ibid. ||
300 (328)
ko ņāma bhaņijja buho ņāum savve paroyaye bhāve majjham iņam ti ya vayaņam jāņamto appayam suddham ||

Page #343

Appendix 4 Samayasāra
329
306 (334)
padikamanam padisaranam parihāro dhāraṇā niyatti ya| ņiņdā garuhā sohi atthaviho hoi visakumbho ||
307 (335)
apaļikkamaņam apadisaranam aparihāro adhāraņā ceva aņiyatti ya aņiņdā agaruhā 'sohi amayakumbho ||
324 (-)
vavahārabhāsieņa u paradavvam mama bhaņamti vidiyatthā || jāņamti ņicchayeņa u ņa ya mama paramāņumettam avi kimci ||
325 (-)
jaha kovi ņaro jampai amhā gāmavisayaņayararatham | ņa ya honti tassa tāņi u bhaņai ya moheņa so appā ||
326 (-)
em eva micchaditthi ņāņi ņissamsayam havai eso jo paradavvam mama idi jāņamto appayam kuņai ||
345 (357)
kehi ci du pajjayehim viņassae ņeva kehi ci du jivo jamhā tamhā kuvvai so vā anno va ņeyamto ||
346 (358)
kehi ci du pajjayehim viņassae ņeva kehi ci du jivo jamhā tamhā vedadi so vā aṇno va ņeyamto ||
347 (359)
jo ceva kuņai so ceva vedako jassa esa siddhamto so jīvo ņāyavvo micchādițhi aņārihado ||
348 (360)
aņņo karei aņņo paribhumjai jassa esa siddhamto | so jīvo ņāyavvo micchādiţthi aņārihado ||

Page #344

330 Appendix 4 Samayasāra
356 (385 SBJ):
jaha sețiyā du ņa parassa sețiyā sețiyā ya sä hoi|| taha jāņao du ņa parassa jāņao jāņao so du ||
390 (420)
sattham ņāņam ņa havai jamhā sattham ņa yāņae kimci | tamhā aņņam ņāņam aņņam sattham jiņā viņti ||
403 (433)
jamhā jāņai ņiccam tamhā jīvo du jāņao ņāņi| ņāņam ca jāņayādo avvādirittam muņeyavvam ||
404 (434)
ņāņam sammādithi du samjamam suttam amgapuvvagayam! dhammādhammam ca tahā pavvajjam abbhuvamti buhā ||
408 (438)
pāsamdiyalimgāņi va* gihalimgāņi va bahuppayārāņi |
ghittum vadamti mūdhā limgam iņam mokkhanaaggo tti || *JGM ed. has pākhamdiya - see Pischel para. 265: pākhamdi is the erroneous writing of kha for șa.
409 (439)
ņa du hoi mokkhamaggo limgam jam dehanimmamā arihā limgam mucittu damsaņaņāņacarittāņi sevamti ||
410 (440)
ņavi esa mokkhamaggo pāsamdīgihamayāņi limgāņi | damsaņaņāņacarittāņi mokkhamaggam jiņā viņti ||
411 (441)
jamhā jahittu linge sāgārañagāraehim vā gahie damsaņaņāņacaritte appāņam jumja mokkhapahe ||
412 (442)
mokkhapahe appāņam thavehi tam ceva jhāhi tam ceva* | tattheva vihara niccam mā viharasu annadavvesu ||

Page #345

Appendix 4 Samayasāra
331
* JGM ed. reads: thavehi vedayadi jhāyahi ceva, which is preferable.
413 (443)
pāsamờilimgesu va gihalimgesu va bahuppayāresu | kuvvamti je mamattam tehim na ņāyam samayasāram ||
414 (444)
vavahārio puņa nao donni vi limgāņi bhaņai mokkhapahe |
ộicchayanao na icchai* mokkhapahe savvalimgāņi || * JGM ed. reads: ņiccayaņao du ņicchadi
415 (445)
jo samayapāhudam iņam pathiūņa atthataccao ņāum* |
atthe țhāhidi ceyā so hohi** uttamam sokkham || * JGM ed. reads: pațhidūņaya acchataccado nādum and **pāvadi, which must be right

Page #346

Apabhramsa
Yogindu
Pāli
Bibliography
Primary Sources and Translations
Dhammapada
Majjhima-nikāya Samyutta-nikāya
Suttanipāta Vinaya-pitaka
Visuddhimagga
Prākrit
Āyāramga Sutta

Paramātmaprakāśa and Yogasāra, ed. A.N. Upadhye, Bombay, 1937.
S. Radhakrishnan (ed. and tr.), Oxford, 1950.
Pali Text Society edition, London. Pali Text Society edition, London. - trans. V. Fausboll [Sacred Books of the East, Vol. X, Pt. 1] Oxford, 1881. Pali Text Society edition, London. Pali Text Society edition, London. - Suttavibhanga tr. I.B. Horner in The Book of the Discipline (VinayaPitaka), Vol. I [Sacred Books of the Buddhists, Vol. X] London, 1938. Pali Text Society edition, London.
in Puppha Bhikkhu (ed.), Suttāgame I, Gurgaon, 1953, pp. 1-99.
H. Jacobi (ed.), The Ayāramga Sutta of the Svetambara Jains, London [Pali Text Society], 1882.
tr. with an intro. by H. Jacobi as 'The Acārānga Sutra', in Jaina Sutras Pt. I, [Sacred Books of the East, Vol.XXII] Delhi, 1980 [reprint of 1st ed., Oxford, 1884], pp. 1-213. M. Jambuvijaya (ed.), Āyāranga

Page #347

Bibliography 333
suttam, Jaina-āgama Series No. 2(1), Bombay, 1976. - W. Schubring (ed.), ĀcārangaSūtra:Erster Sruta Skandha (Text, Analyse und Glossar), original ed. Leipzig, 1910), reprint Liechtenstein, 1966 [Abhandlungen für die Kunde des Morgenlandes herausgegeben von der Deutschen Morgenländischen Gesellschaft,
Band XII, No. 4). Sūyagadamga Sutta in Puppha Bhikkhu (ed.), Suttāgame
I, Gurgaon, 1953, pp. 101-182. - Studien zum Süyagada: die Jainas und die anderen Weltanschauungen vor der Zeitwende. Textteile, Nijjutti, Ubersetzung und Anmerkungen, W.B. Bollée, (Schriftenreihe des Südasien-Instituts der Universität Heidelberg, Band 24] Teil I, Wiesbaden, 1977. - Studien zum Süyagada. Textteile, Nijjutti, Ubersetzung und Anmerkungen, W.B. Bollée, [Schriftenreihe, etc., Band 31] Teil II, Stuttgart, 1988.
M. Jambuvijaya (ed.), Süyagadamga-suttam, Jaina-Agama Series No. 2(2), Bombay, 1978. - tr. with an intro. by H. Jacobi as 'The Sūtrakrtānga Sūtra' in Jaina Sūtras Pt. II, [Sacred Books of the East, Vol. XLV] Delhi, 1973 [reprint of 1st ed., Oxford, 1895), pp. 235
435. Dasaveyāliya Sutta in Phulchandji Mahārāj (ed.),
Suttāgame II, Gurgaon, 1954, pp.

Page #348

334 Bibliography
Uttarajjhayana Sutta in Phulchandji Mahārāj (ed.),
Viyahapannatti (Bhagavai)
Kundakunda (1)
Kundakunda (2)
Kundakunda (3)
947-976.
- tr. by K.C. Lalwani, Delhi, 1973. - E. Leumann (ed.), W. Schubring (tr.), Ahmedabad, 1932.

Suttagame II, Gurgaon, 1954, pp. 977-1060.
- tr. with an intro. by H. Jacobi as 'The Uttaradhyayana Sūtra' in Jaina Sutras Pt. II, [Sacred Books of the East, Vol. XLV] Delhi, 1973 [reprint of 1st ed., Oxford, 1895], pp. 1-232. - J. Charpentier (ed. with an intro., critical notes and a commentary), The Uttaradhyayanasūtra, Uppsala, 1922.
in Puppha Bhikkhu (ed.), Suttāgame I, Gurgaon, 1953, pp. 384-939.
J. Deleu (intro., critical analysis, commentary and indexes), Belgium, 1970.
-
Niyamasara, Prakrit text, ed. and tr., with intro., notes and comm. by Uggar Sain, [Jagmandarlal Jaini Memorial Series, Vol. V] Lucknow, 1931.
- Niyamasāra - reprint of above, critically edited by Padmashri Sumatibhai Shah, Solapur, 1988. Pañcāstikāyasāra, Prākrit text, ed. and tr., with intro., notes and an original comm. in English by A. Chakravartinayanar, [The Sacred Books of the Jainas, Vol. III] Arrah, 1920.
Pravacanasara, Prākrit text, tr. and ed. with an intro. by A.N. Upadhye

Page #349

Bibliography
335
Kundakunda (4)
(includes Sanskrit commentaries of Amộtacandra (Tattvadīpikā) and Jayasena (Tātparya-vrtti), and Pande Hemaraja's Hindi commentary), (2nd ed.) Bombay, 1935. - The Pravacana-sāra of Kundakunda Ācārya, together with the commentary, Tattva-dipikā, by Amrtacandra Sūri, English tr. by B. Faddegon, ed. with an intro. by F.W. Thomas, Cambridge, 1935. Samayasāra, Prākrit text, ed. and tr. with intro. and comm. by A. Chakravarti, Banaras, 1950. - Samayasāra, Prākrit text, tr., comm. and intro. by J.L. Jaini, [Sacred Books of the Jainas, Vol. VIII] Lucknow, 1930. - Samayaprābhrtam, G. Jain (ed.) [includes Sanskrit commentaries of Amộtacandra (Ātmakhyāti) and Jayasena (Tātparya-vrtti)], (Sanatana Jaina Grantha Mala 3) Benares, 1914.
Sanskrit
Amstacandra
Bhagavadgitā Jayasena
- see Kundakunda (3), Tattvadīpikā. - see Kundakunda (4), Ātmakhyāti. R.C. Zaehner (ed.), Oxford, 1969. - see Kundakunda (3), comm. on Pravacanasāra. - see Kundakunda (4), comm. on Samayasāra. Patañjala Darshana (Yoga Sūtras), (with the commentary of Vyāsa), J. Vidyasagara (ed.), Calcutta, 1940.
Patañjali

Page #350

336
Bibliography
Pūjyapāda (Devanandin)
Umāsvāti
Sarvārthasiddhi, P.S. Shastry (ed.), [Sanskrit Grantha No. 13] Banaras, 1955. - Reality, tr. of Sarvārthasiddhi by S.A. Jain, Calcutta, 1960. Tattvārthādhigama Sūtra, ed. and tr. by J.L. Jaini, [Sacred Books of the Jains, Vol. II] Arrah, 1920. - Tattvārthādhigama (with bhāsya), ed.. K.P. Mody, Calcutta, 1903. Tattvārtha Sūtra, ed. Pt. Sukhlalji with his comm. and tr., with an intro. by K.K. Dixit, [L.D. Series 44) Ahmedabad, 1974. The Nyāya-Sūtras with Vätsyāyana's Bhāsya, ed. G.S. Tailanga [Vizianagram Sanskrit Series, Vol. IX] Benares, 1896. - Gautama's Nyāyasūtras (With Vātsyāyana-Bhāsya), tr. G. Jha, Poona, 1939.
Vātsyāyana
Dictionaries and Reference Works
Ardha-māgadhi koșa
Buddhist Dictionary
A Critical Pāli Dictionary
Ratnachandra (ed.), intro. by A.C. Woolner, Vol. I, Ajmer, 1923. compiled by Nyanatiloka, 4th revised edition, Kandy, 1980. begun by V. Trenckner, revised, continued and edited by D. Andersen, H. Smith and H. Hendricksen, Vol. I, Copenhagen, 1924-1948.
A Grammar of the Prākrit Languages
R. Pischel, tr. from
the

Page #351

Paia-Sadda-Mahannavo: A Prakrit-Hindi
Dictionary
A Sanskrit-English Dictionary
Bibliography 337
German by S. Jha, 2nd revised edition, Delhi, 1981.
compiled by H.D.T. Sheth, Calcutta, 1928.
M. Monier-Williams, Oxford, 1979 (reprint).
Works Cited
(excluding primary sources in classical languages and dictionaries, for which see above)
Alsdorf, L.
1958 'Itthiparinnā: A Chapter of Jain Monastic Poetry, Edited as a Contribution to Indian Prosody', Indo-Iranian Journal, Vol. II, The Hague, pp. 249-270.
1962- 'Uttarajjhāyā Studies', Indo-Iranian Journal, 1963 Vol. VI, The Hague, pp. 110-136. 1965 Les Études Jaina, College de France. 1966 The Arya Stanzas of the Uttarajjhāyā, Wiesbaden.

1977 'Jaina Exegetical Literature', in A. N. Upadhye, et al. (ed.), Mahāvīra and His Teachings, Bombay, pp. 1-8.
Basham, A.L.
1951 History and Doctrine of the Ājivikas, London. Bhatt, B.
1974
'Vyavahāranaya and Niścayanaya in Kundakunda's Works', Zeitschrift der Deutschen Morgenländischen Gesellschaft, Suppl. 2, Leipzig, pp. 279-291.

Page #352

338 Bibliography
Bhargava, D.
1968 Jaina Ethics, Delhi.
Bollée, W.B.
1977 Studien zum Suyagaḍa: die Jainas und die anderen Weltanschauungen vor der Zeitwende. Textteile, Nijjutti, Ubersetzung und Anmerkungen [Schriftenreihe des SüdasienInstituts der Universität Heidelberg, Band 24], Teil I, Wiesbaden.
1988 Studien zum Suyagaḍa.
Bronkhorst, J.
Textteile, Nijjutti, Übersetzung und Anmerkungen [Schriftenreihe, etc., Band 31], Teil II, Stuttgart.
1985 'On the Chronology of the Tattvärtha and some Early Commentaries', Wiener Zeitschrift für die Kunde Südasiens, Band XXIX, Vienna, pp. 155184.
Caillat, C.
1974 'Jainism', in C. Caillat, A.N. Upadhye, B. Patil (ed.), Jainism, Bombay, pp. 1-48.
1975 Atonements in the Ancient Ritual of the Jaina Monks, Ahmedabad.
1987 'Jainism', in M. Eliade (ed. in chief), The Encyclopedia of Religion, Vol. 7, New York, pp. 507-514.
Chatterjee, A.K.
1978 A Comprehensive History of Jainism, Vol. I, Calcutta.
Collins, S.
1982 Selfless Persons, Cambridge.
Deleu, J.
1970 Viyahapannatti (Bhagavai): Introduction, Critical Analysis, Commentary and Indexes, Belgium.
1977 'Lord Mahāvīra and the Anyatirthakas', in A.N. Upadhye et al. (ed.), Mahāvīra and His Teachings, Bombay, pp. 187-193.

Page #353

Bibliography 339
Deo, S.B.
1956 History of Jaina Monachism, Poona. Dixit, K.K.
1971 Jaina Ontology, [L.D. Series 31] Ahmedabad.
1978 Early Jainism, [L.D. Series 64] Ahmedabad. Doshi, S.
1985 Masterpieces of Jaina Painting, Bombay. Eliade, M. 1988 'Yoga', in M. Eliade (ed. in chief), The
Encyclopedia of Religion, Vol. 15, New York,
pp. 519-523. Frauwallner, E. 1953 Geschichte der Indischen Philosophie, Band 1
[Band 2, 1956], Salzburg. 1973 History of Indian Philosophy, (2 Vols.), trans.
V.M. Bedekar, Delhi. Glasenapp, H. von 1942 The Doctrine of Karma in Jain Philosophy,
trans. G.B. Gifford, Bombay. Gombrich, R.F. 1971 Precept and Practice: Traditional Buddhism in
the Rural Highlands of Ceylon, Oxford. 1984 'Notes on the Brahmanical Background to
Buddhist Ethics' in G. Dhammapala, R. Gombrich, K.R. Norman (ed.), Buddhist Studies in Honour of Hammalava Saddhatissa, Sri
Lanka, pp. 91-102. 1988 Theravāda Buddhism: A Social History from
Ancient Benares to Modern Colombo, London. Jacobi, H.
1884 See Āyāramga Sutta.
1895 See Sūyagadamga Sutta / Uttarajjhayaņa Sutta. Jaini, J.L.
1940 Outlines of Jainism, Cambridge. " Jaini, P.S. 1977 'Bhavyatva and Abhavyatva: A Jain Doctrine of
"Predestination", in A.N. Upadhye, et al. (ed.),

,

Page #354

340 Bibliography
Mahāvīra and His Teachings, Bombay, pp. 95
111. 1979 The Jaina Path of Purification, Delhi. 1980 'Karma and the Problem of Rebirth in Jainism',
in W. O'Flaherty (ed.), Karma and Rebirth in Classical Indian Traditions, California, pp. 217
238. Jain, S.A. 1960 Reality: (English translation of Shri
Pūjyapāda's Sarvārthasiddhi), Calcutta. Johnston, E.H. 1974 Early Sāņkhya, Delhi (reprint of original
edition, 1937, London). Keith, A.B. 1936 Review of A.N. Upadhye's edition of the
Pravacanasāra in The Journal of the Royal
Asiatic Society, London, pp. 528-9. Lamotte, E. 1988 History of Indian Buddhism: From the Origins
of the Saka Era, trans. S. Webb-Boin, Paris
(original French edition, 1958, Paris). Matilal, B.K. 1981 Central Philosophy of Jainism (Anekānta Vāda),
[L.D. Series 79] Ahmedabad. 1986 The Logical Illumination of Indian Mysticism',
in N.J. Allen, R.F. Gombrich, T. Raychaudri, G. Rizvi (ed.), Oxford University Papers on India,
Vol. 1, Pt. 1, Oxford. Norman, K.R. 1977 "Kāvilīyam: A Metrical Analysis of the Eighth
Chapter of the Uttarādhyayana-Sūtra', in A.N. Upadhye et al. (ed.), Mahāvīra and His
Teachings, Bombay. Ohira, S. 1982 A Study of Tattvārtha Sūtra with Bhāsya, (L.D.
Series 86] Ahmedabad.

Page #355

Bibliography 341
Olivelle, P.
1974 The Origin and Early Development of Buddhist Monachism, Ceylon.
Pischel, R.
1981 A Grammar of the Prakrit Languages, tr. from the German by S. Jha, 2nd revised edition, Delhi, 1981.
Poussin, La Vallée
1923- L'Abhidharma de Vasubandhu, IV, Paris. 1931
Puligandla, R.
1975 Fundamentals of Indian Philosophy, New York. Rahula, W.
1978
What the Buddha Taught, [2nd edition] London and Bedford.
Rice, B.L.
1914 Epigraphia Carnatica I, Coorg Inscriptions No. 1, [revised ed. 1972], Madras. Sangave, V.A.
1980 Jaina Community: A Social Survey, Bombay. Schubring, W.
1935 Die Lehre der Jainas, Berlin and Leipzig. 1957 'Kundakunda echt und unecht', ZDMG, Vol. 107, Leipzig, pp. 537-574.
1962 The Doctrine of the Jainas, trans. W. Beurlen [of Die Lehre der Jainas], Delhi [reprint 1978]. 1966 The Religion of the Jainas, Calcutta. Singh, R.J.
1974 The Jaina Concept of Omniscience, [L.D. Series 43] Ahmedabad.
Tatia, N.
1951 Studies in Jaina Philosophy, Ahmedabad. Tieken, H.

1986 'Textual Problems in an Early Canonical Jaina Text', Wiener Zeitschrift für die Kunde Südasiens, Band XXX, Vienna, pp. 5-25.

Page #356

342
Bibliography
Upadhye, A.N. 1935 'Introduction' to Pravacanasāra (see
Kundakunda (3)], Bombay, pp. i-cxxvi. Williams, P. 1989 Mahāyāna Buddhism: The Doctrinal
Foundations, London. Williams, R. 1963 Jaina Yoga: A Survey of the Mediaeval
Śrāvakācāras, London. 1966 "Before Mahāvīra', Journal of the Royal Asiatic
Society, London, pp. 2-6. Zydenbos, R.J. 1983 Moksa in Jainism, According to Umāsvāti,
[Beiträge zur Südasienforschung, Band 53] Wiesbaden

Page #357

Glossary and Index
Alsdorf, L., 4, 14 Amrtacandra, 94, 113-114,
116, 129-130, 133, 157, 160161, 164, 168, 170-171, 173, 175, 179-180, 203, 206-207, 220, 228, 231, 233, 236, 240-242, 250, 259, 289n,
296 anagāra - homeless (ascetic),
297 anaśana - fasting, 223 anekānta(vāda) - (doctrine of) manifold aspects, 103, 137, 183n, 232, 244, 247-248,
251-255, 258, 262-265, 305 anubhāva - fruition of karma,
55
abhavya - souls incapable of
attaining liberation, 245-246 ācārya - religious teacher, 260 acetanā - non-consciousness,
262 acetanā - non-conscious
substance, 262 activity (of body, speech and mind), 7-12, 15-16; in the Tattvārtha Sūtra, 47-90; see
also yoga ādāna-niksepana-samiti - care
in picking things up and
putting them down, 69-70 adharma - demerit, 289 adhikaraṇa - substratum (of
karmic influx); of jiva type, 64-66; of ajiva type, 67-72;
68, 209 adhyavasāna - intention,
resolution, will, 272-274;
synonyms for, 273, 275, 294 Adipurāņa, 89 ahimsā - non-injury, non
violence, 1, 11, 20, 25, 63, 73, 81, 83-84, 89, 156, 158, 166, 169, 177-179, 182, 189, 221, 230, 264, 290, 294-295,
303 ; ajīva - insentient, 64, 67-68,
103, 120n, 129, 139, 142, 146, 148, 150, 161, 170, 173, 205, 212, 215, 217, 243, 255, 257, 268, 271, 283, 285,
290; five ajiva-dravya, 275 . Ājivikas, 41 ajñāna - ignorance, 111, 268 akusala - karmically
unwholesome, 13 alms, 27-29, 222-223
anukampā - see compassion anuprekşā - reflection, 192,
196 anuvrata - partial vows (taken
by laity), 73-74, 74n, 78; see
also mahāvrata Apabhramsa, 302; use of, 93,
95 aparigraha - non-possession,
non-possessiveness, non-. attachment (fifth aņuvrata of lay person), 74-75, 74n, 88,
169, 219, 289-290 appamāda - mindfulness, 52 ārambha - 'undertaking',
premeditated action, violence, 5-6, 26, 31-33; meaning of, 38-40; 65-66,.. 66n, 72, 154, 173-174, 220,
311 arhat - 'worthy of worship', omniscient being, Jina, 35,
172, 210, 211, 247, 295, 297 asat - unreal, 142, 148
!

Page #358

344 Harmless Souls
asava - four cankers, corruptions, 14; see also āsrava ascetic (Jain), 20-24, 27, 33, 37, 63, 72, 76-78, 80-89, 11112, 115, 117, 163, 165-168, 189, 195, 198, 217-230,271, 283; characteristic marks of, 295-303; ascetic practice, 217-224, 295-310; see also monks, nuns
āśrama - stage of life (in orthodox Hinduism), 89 äsrava influx of karmic matter, 10, 14-15, 25, 48-49, 56, 58-59, 62-63, 65, 71-72, 86, 163, 242, 281, 285; see also āsava
āsravadvāra - cause of asrava,
-
56
āśraya - substratum, base, 104105
aśubha-upayoga - inauspicious manifestation of
consciousness, 98, 106n, 108-109, 112, 114-115, 118, 120, 122, 133-134, 139, 143, 145, 219, 225-226 aśubha-yoga-wicked
(inauspicious) activity, 49, 49n, 138
aśuddha-upayoga - impure
manifestation of
consciousness, 108, 112-115, 117, 120, 122, 138, 141, 146, 154, 156, 158-159, 161-162, 170, 172-177, 180, 182-183, 205, 219, 225-226, 306 Ātmakhyāti, 96, 231, 236-237,
248-249, 289n, 297-298 ātman - self, 112-114, 118
122, 124, 132, 135-136, 138139, 142n, 143, 146-148, 170, 181, 183, 195, 203-204,

209, 211, 215, 217, 226, 238, 243, 246, 257, 260-261, 280285, 298, 302; ātmanbrahman, 126; incorporeal (amūrta), 128-129; paramātman, 139-140, 148, 278; śuddhātman, 174, 228 Atthapähuda, 94-95 atthikāya (asti-kāya) - fundamental entity, 99 avaśyaka - obligatory duty of monk, 43n, 191, 193n Avasyaka-niryukti, 190 avidya- ignorance, 111, 142, 149-150, 154-156,; avijjā, 151n
avirati - non-restraint; cause of bondage, 54-56, 58
avrata - non-observance of the five vows (vrata), 58; see also vrata
avyākta - karmically indeterminate acts (Buddhist), 13 Ayāramga Sutta, 4-5, 21-24, 26, 31, 34-35, 39, 43-44, 84, 97, 177-178, 188-189 ayuş-karma - age-determining karma, 293
B bandha - bondage, 56, 129 Bārasa-Anuvekkhā, 92-94 Bhagavai (Viyahapannatti) - see Viyahapannatti Bhagavadgită, 228, 292-293 bhakti - devotion, 232 Bharata, 93
bhāṣā-samiti - care in
speaking, 69-70
Bhatt, B., 243, 255-256, 263, 265
bhāvu - disposition, mental state, nature, state of the

Page #359

Glossary and Index 345
164, 167, 180, 232, 265, 273, 285, 296, 299; attitude to karmic bondage, 11-17, 30, 38, 42, 52
C
soul; five kinds, 103-105; 107-109, 118, 131, 132, 186, 273, 275; bhāva-karman, 143; aśubha-bhāva, 107109, 118; jīva-bhāva, 99, 104; parama (suddha)-bhāva - liberated self, 267; parināmika-bhāva, 106-107, 106n, 116; renunciation of, 288-295, 305; in Samayasāra, 231, 249, 257261, 267-273; śubha-bhāva, 107, svabhāva -own-nature, essence, 118, 203, 209-210, 234-235, 239, 244n, 276, 279, 281, 284-286, 288, 294
295, 297 bhāvabandha - efficient cause
of bondage, 107, 131, 133.
134 bhāvanā - mental disposition, mental development, 287288; ātmabhāvanā -
meditation on the self, 288 Bhikhanji, 183n-184n bhikkhu - Buddhist monk, 167 bhūtārtha - really existent,
truth, 242-243, 257 bondage - see karmic bondage brahman - Upanisadic and
Vedāntic absolute, 238; see
also ātman Brahmanical ritual, 9-10, 77 Brhadāranyaka Upanişad, 120 Bronkhorst, J., 47 Buddha (Gotama), 9, 12-13,
180n Buddhism, 164, 284; Ch'an,
301; early, 38, 283; Mādhyamika, 137, 249-250, 252-253; Mahāyāna, 180, 282, 285-286; Theravāda,
208 Buddhists, 96, 135, 154-155,
Caillat, C., 14, 18-20, 233n caitanya - consciousness, 101,
104-105, 170; see also
cetanõn carefulness, 52, 52n, 161, 165 carelessness, 42-43, 160-161,
166; see also pramāda caritra - conduct, 118-119,
196, 234-235, 298, 306; in the Pravacanasāra, 185-187;
sāmāyika-cāritra, 188-193 cāritra-mohaniya karma -
right-conduct deluding
karma, 56-57, 149, 150-151 cetanā - consciousness, 125
127, 262; volition, 13; cetana-dravya - conscious
substance, 262 Chakravarti, A., 237, 278, 301 Chatterjee, A.K., 93-94 cheda - infringement, offence
against the monastic rules, 157-160, 162, 171, 173; antaranga (internal) cheda, 160-161; bahiranga
(external) cheda, 160-161 Chedasūtras - see Cheya Suttas Cheya Suttas, 19, 85 compassion, 10, 17, 175-184 consciousness - see upayoga,
caitanya, etc.
D damsana - see darśana daņda - inflictions, violence,
punishment, 12-13 darśanal- perception, insight,
faith, 98-109, 112-113, 140,

Page #360

346 Harmless Souls
186, 209, 234-235, 257, 298; four kinds of, 105-108; see also upayoga
darśana2- religion, school of thought, 80, 124 darśana-mohaniya karma - perception-deluding karma,
149-152
Dasaveyāliya Sutta, 4, 8, 1011, 20-22, 24-26, 28-29, 31, 33, 35-37, 43n, 72, 97, 157 Deleu, J., 40-42, 100-101 demerit (papa), 98, 115-116, 138, 143, 226, 272, 287, 294; papa-karma, 48-49, 49n, 88
Devanandin - see Pujyapāda dharma (dhamma)1 - teaching, law; Jain, 27-28, 169, 186187, 192-193, 196, 214; defined in the Pravacanasāra, 118-119; parama-dharma - highest dharma, 170; svadharma, 228
dharma2- principle of motion,
289
dharma3- merit, 289 dhyāna (jhāna) - meditation, concentration, 153, 156, 186-188, 196-205, 208, 286, 305; four types listed, 198200; dharma-dhyāna, 194; śukla-dhyāna, 194 Digambara - 'sky-clad' Jaina
sect, 265, 309, passim; cf. Śvetāmbara
Dixit, K.K., 5-9, 23-24, 29, 3132, 35, 61, 144, 244 doctrinal development, 80-90,
128
dosa - hatred, aversion, 13, 138, 154-155; see also dveṣa

dravya-substance, 104, 129,
131, 209, 212, 247, 254, 267, 276, 294; listed, 289; acetana-dravyaunconscious substance, 204 dravya-bandha - material cause of bondage, 107, 131, 133-134
Dravya-samgraha - 126-127 duḥkha, - suffering, 98, 119, 138 dveșa-hatred, aversion, 129,
132-135, 141, 145, 150, 152155, 180; see also dosa dvikriyāvāda (dokiriyāvāda) - doctrine that one cause can produce two different effects, 260-266
E
ekāgra - one-pointed thought, concentration, 207, 210 ekanta - one-sided, 244, 247, 253, 264; ekāntavädin - one who has a single view of reality, absolutist, 244; cf. anekānta
ekendriya - single-sensed beings, 17-18, 26; see also sthāvara
Eliade, M., 207
Eliot, T.S., 164 emblem - see linga
eṣaṇa- desire, alms, 222-223 eṣaṇā-samiti - care in
accepting alms, 69-70, 223 evil (violent) act, 7-11, 20 evil-doers, 6; behaviour of, 711
F
Faddegon, B., 155, 157, 171, 179, 215 fatalism, 293

Page #361

Glossary and Index
347
Frauwallner, E., 79, 233, 240
gati - birth (four possible), 23 gnosis, 122, 141, 153, 156,
264, 272, 307 Gombrich, R.F., 28n, 29n, 36,
164-165, 238 guna - essential property,
quality, 99, 101, 104, 114,
126, 254, 276 Guņacandra, 94 gunasthāna - fourteen stages
on path to liberation, 83-84,
83n, 88-89, 199 gupti - three kinds of restraint. 44, 69n, 71, 191-192, 194, 196, 203
279 Indra, 177 indriya - sense(s), 58 intention - see volition internalisation (of doctrine,
practice), 303-310, passim iryāpatha - four postures (in
Buddhism), 62 īryāpatha āsrava - short-term
karmic influx, 56, 59-60 īryāpatha karma - short-term karma, 48, 49n, 50-51, 65, 86-87; in the Tattvārtha
Sutra, 57-64 īryāpathakriyā - walking
carefully, 59-62 īryā-samiti (iriyā-samii) - care
in walking, 41-44, 61-62, 69
70, 168-169; see also samiti īriyāvahiyā kiriyā - action of a
monk conforming to the monastic rules, 41, 43n, 44; cf. samparāiyā kiriyā
Sun.acanda 276
H Hardy, F., 233n hātā - relinquisher, 143 hiņsā - violence, harming
action, injury, 1, 11 19-22, 25-26, 29, 35, 37, 39, 42, 5152, 54, 62-63, 65, 68, 72, 7577, 152, 202, 219, 292, 294, 306; defined in the Tattvārtha Sūtra, 51-52; in
the Pravacanasāra, 156-184 Hindus, 228, 232, 297, 303 householders, 5-6, 32, 32n, 37
38, 63-64, 72-78, 80-89, 11112, 115, 168, 198, 26-230, 282-283,304-310; characteristic marks of, 295303; practice of sāmāyika, 189-190; status in earliest texts, 23-31
Jacobi, H., 12, 32, 44 Jaini, J.L., 231-238, 258, 300 Jaini, P.S., 5, 9, 14-15, 46, 54, 73-74, 85-86, 101, 104, 121, 128, 131, 149-151, 183n,
188-191, 195, 211-212, 254 Jayasena, 94, 96-97, 168, 171,
176, 231, 233, 236, 242,
257-260, 287, 296 jina - spiritual conqueror, 213
214, 235, 261, 275, 282-283,
295 Jinasena, 89 jiva - soul, life-monad, sentient being, 1, 6, 10, 17, 18, 2426, 36, 38, 48, 50-51, 61, 6364, 67-68, 72, 75, 81, 87, 98, 101-106; 109-110, 113, 116118, 122, 124-131, 133-140,
identity, socio-religious,
passim impurity, 77, 136, 205-206,

Page #362

348 Harmless Souls
142n, 143, 146-147, 149150, 152, 154, 163, 165, 169-170, 181, 186, 200-202, 205-206, 212, 215, 234-235, 243, 248, 255, 257, 261-263, 267-269, 271-276, 280-281, 283-285, 292; involvement with karma, 124-149; qualities of, 101-106 jñāna - knowledge, realisation, 98-109, 112-113, 117, 140, 156, 167, 185, 187, 192, 198, 200, 209, 212-217, 234-235, 257, 275-278, 291-292, 298, 303, 305; eight kinds of, 105-109; jñāna-cetanā, 195; matijñāna, śrutajñāna, 102; see also upayoga
Johnston, E.H., 95
K
kamma (Pāli) - action, deed, 12 karma(n) - action, matter
which binds the soul, 16-17, 20-21, 86, 105, 106n, 107, 109, 111, 115-116, 120, 122, 124-125, 128, 130, 134-135, 137, 139-145, 150-151, 154, 163-164, 182, 185, 196, 198, 201-203, 206, 208, 211-212, 217, 236, 242, 255-257, 261, 268-271, 275, 281, 285, 291, 293-295, 300-301, 303-305; dematerialisation of, 145149; dravya-karma, 135, 144; karma-pudgala, 129, 132-137, 142
karmic bondage, 1-2, 11-21, 33, 37, 64, 75, 77, 81, 85; five causes of, 54-55, 58; in the Pravacanas āra, 124-184, 212; in the Samayasara, 267-274; short-term, 41-44; in the Tattvärtha Sutra, 46

90
kartā - agent, 143, 261 karuṇā (kaluna), 177; see compassion
kaṣāya - 'stain', passion, four moral vices, 13, 18, 31-38, 44, 48, 49, 50-51, 55-59, 61, 63, 65-66, 68, 72-78, 83, 86-88, 107-108, 110, 117, 130, 150-154, 169, 175, 268n, 273, 291-292; see also passion, krodha (anger), māna (pride), māyā (deceit), lobha (greed) Kaṣāyaprabhṛta, 127 Katha Upanisad, 292 kaya-danda - infliction of the body, 12
kāya-samrambha - bodily impulsion, 65-66 kāyotsarga- abandonment of
the body, meditative posture, standing motionless, 39, 190 kevaladarśana - perception
without karmic obstruction, 106-109
kevalajñāna - omniscience, knowledge without karmic obstruction, liberation, 103, 106-109, 116, 181, 209, 211212, 217, 276-278, 281-281, 297, 299
kevalin - one who has attained omniscience and liberation, 147n, 213, 215, 279, 281282; see also siddha kiriyā - action; types of, 40-41, 44, 58; see also kriyā, iriyāvahiyā kiriyā,
samparāiya kiriyā knowledge - see jñāna kriyā - action, 58-61, 61n; see also kiriyā
krodha - anger, 34-35, 66, 150,

AI THỊ ĐÔI THẾ KHÔNG T

Page #363

Glossary and Index
349
152-153; see also kasaya Kundakunda, 52, 86, 89-91,
91-310 passim; date of 91-98 kusala - karmically
wholesome, 13
labdhi - dormant
consciousness, 102 lakkhana - see lakşaņa laksaņa - distinguishing
characteristic, 99, 102;
višeșa-laksana, 113 lay Jains - see householders leśyā - colour (of soul), 130 liberation, 182, 184-185, 215
216, 245, 264-265; in Samayasāra, 275-310; see also mokşa linga - emblem, defining
characteristic (of ascetic or householder, 217-222, 295.
303, 309-310 lobha - greed, 13, 34-37, 66,
72, 75-77, 150, 152-154, 219; see also kaşāya lokākāśa - inhabited universe,
132
material soul, 129-130, 143,
262-263, 304 Matilal, B.K., 244n māyā - deceit, 34-35, 66, 150,
152-153; see also kasāya mechanism of bondage - see
karmic bondage meditation, 140, 147, 181, 197
200, 286-288, 303; in
Pravacanasāra, 200-217 mendicants - see ascetics,
śramanas merit (punya), 26-31, 48-49,
49n, 81, 88, 98, 115-116, 138, 143, 226, 272, 280, 287, 293-294; punya-karma, 30,
273 Merkara (Mercara) copper
plates, 92-94 mithyā-darśana - wrong-belief
(cause of bondage), 54-56,
58, 150 mithyātva (karman) -
perversity, 150-151 moha - delusion, false
consciousness, 13, 119, 122, 129, 132, 134-135, 138-142, 145-156, 169-170, 176, 179180, 209, 268, 270, 289;
darśana moha, 150, 156 mohaniya karma - deluding
karma, 149, diagram 150,
151, 155-156, 205 mokṣa - liberation, 23, 72, 82,
115, 122, 141, 183n, 185, 198, 278-279, 298; see also
liberation Monier-Williams, M., 296-298 monk (Jaina); behaviour of,
18-26, 37, 63-64, 72, 76-77, 81, 83, 157, 164, 168, 176; monastic rules, 41-44, 158
172; see also ascetic muhūrta - period of 48
M
madhyastha - neutral position,
251 Mahāvira (Nigantha Nātaputta), 12, 15-16, 41-42,
79-80, 99, 188, 190-191 mahāvrata - major vow (five
taken by ascetics), 31, 73, 78, 84, 169, listed 189, 195, 273, 287; see also aņuvrata Majjhima Nikaya, 12, 15 Malamoud, C., 238 māna - pride, 34-35, 66, 150,
152-153; see also kaşāya manas - mind, 71, 104

Page #364

350 Harmless Souls
minutes, 189, 198 mula - roots (unwholesome), 154
muni - ascetic, mendicant, 245, 250, 273
mūrcchā - infatuation, the delusion of possession, 7374, 152, 169, 173-174, 219220; see also parigraha murta - the embodied, material objects, 124, 128
N
Nāgārjuna, 250, 253, 285 ṇāņa - see jñāna Natyaśāstra, 93
naya - view, aspect of truth, 123n, 127, 136-137, 149, 224, 305; śuddha-naya, 242, 246, 248
Nemicandra, 126-127
Nigantha Nātaputta - see Mahāvira; see also nirgrantha
nigoda - submicroscopic and lowest form of life, 163 nikṣepa - placing, putting down, 67, 69-71
nimitta - instrumental cause,
261
nirgrantha - unattached,
without possessions; name of Mahavira and the Jain community, 19
nirjarā - shedding of karmic matter, 134, 156, 196 nirvāņa - final liberation, 108, 119, 186-187, 228, 236, 245, 282, 286
nirvartană - production,
performance, 67, 69-71 nisarga - movement, urging, operating, 67-71 niścaya-naya - higher view,

non-conventional (absolute) view, 95, 109, 120, 121n, 122n, 123n, 125, 126n, 127, 130 133-135, 137-138, 142144, 146, 201-212, 206, 212, 216-217, 224, 232, 239-266, 269, 272-274, 279, 291-292, 294, 297, 302-303, 305 Niyamasara, 188, 193-195, 212, 215, 251-252, 258-259, 279 no-kaṣāya - subsidiary
passions, 57, 74, 74n, 150 nuns, 64, 72, 76-77, 81, 168; see also ascetic
Nyāyabhāṣya, 233
O
Ohira, S., 49, 49n, 58, 67 omniscience - see kevalajñāna
P
Pāli Canon, 154, 180n Pañcāstikāya, 95, 101-102, 105, 107, 187, 234, 238 paññā - wisdom, 180 Pannavaṇā Sutta, 99 pāpa - see demerit para - other (than the self), 270, 274, 280
paradravya- other substance,
142, 148, 173-174, 181, 183, 203-206, 243, 281, 298 paramārtha - reality, 279-280, 297
paramartha-satya - higher truth (= liberation) in Madhyamika Buddhism, 250, 285
Paramātmaprakāśa, 302 parigraha - possession,
attachment to possessions, possessiveness, 5-6; objects of, 6; 31-38, 72-78, 152,

Page #365

Glossary and Index
351
193 Pravacanasāra, 52-54,91-230,
231, 232, 239, 257, 267, 286,
296, 305-306 pudgala - matter, 67, 71, 125,
128, 131, 138, 143, 201, 204,
257-261, 268, 271, 275 Pujyapāda (Devanandin), 46,
49, 52-54, 92, 160, 189, 192, 196; see also
Sarvārthasiddhi punya - see merit purity, 76-77, 114, 117, 128,
136, 160, 164-165, (of mind) 173, 176-177, 179, 206, 217220, 224-225, 279, 282, 286,
306-310; see also śauca · Puruşārtha-siddy-upāya, 250 pūrvas - 'original' Jain canon,
subsequently lost, 200, 277278
154, 169, 171, 174, 176,
219-220 pariņāma - modification,
transformation, 95, 104, 113, 118, 135, 144, 170, 254, 261,
273 parīşahajaya - victory over the
22 afflictions, 192, 196 paryāya - mode, modification,
104, 132, 212, 246-247, 254,
267 pasamdiya (pāșanda) - ascetic,
295-297 passion, 11, 13, 18, 44, 51, 54
55, 57, 65, 72; see also
kaşaya Patañjali, 207 pațicca-samuppāda - dependent origination, 135,
151n Pischel, R., 93 pradeśa (paesa) - space-points,
55, 100, 132-134 prajñā - wisdom, direct
realisation of truth, 256, 284
286 prajñāpāramitā literature -
Buddhist wisdom literature,
285 Prākrit, use of, 93 prakrti - matter (karmic), 55 pramāda - carelessness,
negligence, 41, 52, 54-56,
58, 152 pramāts - knower, 101 prāņā - life-essentials,
animating principles, 202
202 pratimā - 11 stages of
renunciation for a lay person,
78, 78n, 83, 85-86, 88 pratikramaņa - ritualised
confession of faults, 299-302 pratyākhyāna - renunciation,
R rāga - passion, attachment,
desire, 118-119, 129, 131135, 137, 141, 143, 145, 150, 152-155, 180, 228, 268, 291
292 Rahula, W., 180, 283 ratnatraya - see three jewels rebirth, 23-31, 73, 81, 119,
170; see also gati, samsāra renunciation - passim ritualisation (of ascetic
conduct), 86, 158-159, 163,
166, 182, 308-310; passim roles (socio-religious in the
Pravacanasāra, 224-230 rūksa - dry (of matter), 128
129, 131 rupa - form, 124, 128-129, 133
sādhu - Jain ascetic,

Page #366

352 Harmless Souls
mendicant, 240 sāgāra-uvaoga - see upayoga Sain, U., 194 sallekhanā - ritual death by
fasting, 84 sama (sama) - equanimity,
118-119, 186, 188, 193, 195 samādhi - deep meditation,
193-195, 207-208 samārambha - preparation to
do vilence, 66, 66n, 311 samaya - essential or realised
self, true definition, (doctrine, religion), 95, 192, defined 233-239, 252, 265,
289, 298 Samayasāra, 91, 95-97, 109,
123n, 124, 130, 137, 144, 147, 156, 167, 211, 216,
230-310 sāmāyika - total physical
restraint, attaining
equanimity, 187-196, 208 samiti - 5 rules of conduct,
types of carefulness, 44, 62, 68, listed 69-71, 163, 192,
196; see also īryā-samiti samjvalana - smouldering
passions, 54 Sāmkhya, 95, 232, 263, 292,
294, 303 samkleśa - defilement, 138 sammāditthi - right
understanding, 283 samparāiyā kiriyā - action
characteristic of a lay person,
samrambha - impulsion, determination to do violence,
66, 66n, table samsāra - cycle of rebirth, worlds in which the soul is bound by karma, 1, 23-24, 35, 38, 50, 59, 81, 108, 121122, 135, 137, 142n, 147, 183. 201. 226. 243, 244n. 273, 280, 282, 290, 300; see
also gati samskāra - life-crisis ritual, 89 samvara - stoppage of karmic
influx, 62, 69, 69n, 76, 134,
156, 192, 196, 202 samyag-darśana - right
faith/belief/view, 185 samyag-drsti - right belief,
242, 277, 283, 285 samyag-jñāna - right
knowledge, 185-186 samyak-cāritra - right conduct,
185-186 samyama - self-control,
restraint, 119, 188 samyoga - combining, mixing
up, 67, 69-71 Sankara, 96, 154 Sanskrit, use of 80, 93 sapta-bhangi-naya - sevenfold
predication, 247 Sarvārthasiddhi, 46-47, 50-53,
57, 60-61, 64-65, 67, 71, 7374, 76-77, 88, 92, 102, 150, 152-153, 160, 168, 174-175, 182, 185, 192, 196-200, 218, 220; see also Pūjyapāda
(Devanandin) śāstra - texts, 275 sat - existent, 104 Şatkhandāgama, 126 śauca - purity, complete
freedom from greed, 76-77, 218
00
41
sāmparāyika āsrava - rebirth
causing karmic influx, 56,
59-61 sāmparāyika karma - rebirth
causing karma, 48, 49n, 50, in the Tattvārtha Sūtra 5764, 86-87

Page #367

Glossary and Index 353
122, 138, 156, 158-159, 218
219, 225, 229, 267, 309 sukha - happiness, bliss, 98,
101, 119 Sukhlalji, Pt., 59-61, 64 Süyagadamga Sutta, 4-6, 15
19, 26-28, 30-35, 44, 97 svayambhū (sayambhu) - self
existent, 120 Śvetāmbara - 'white-clad' Jaina
sect, passim; cf. Digambara Śvetāmbara Canon, 2, dating of
4-5, 177 syādvāda - doctrine of
qualified assertion, 232, 244, 247-248, 251, 253, 262, 264, 305
T
Śauraseni (Jaina), 93 scholastics, 304-305 Schubring, W., 4, 9, 94-95,
100, 191 self-modification, 145 siddha - liberated soul, 211;
see also kevalin sila - morality, 164 Singh, R.J., 238 skandha fkhamdha) - molecules, 124-125; see also
pudgala snigdha - sticky, moist (of
matter), 128-129, 131 Somila, 99 sparśa - touch, 131, 134 śramaņa - ascetic, 31, 32n, 11,
115-117, 119, 120n, 134, 140, 157-158, 160-162, 166168, 172, 176-178, 190, 219
220, 225-230, 306-310 śrauta brahman - brahman
qualified to perform the
higher sacrifices, 77 śrāvakācāra - texts dealing
with lay discipline, 190 Sthānakavāsi (Svetāmbara sub
sect), 183n sthāvara - static jīva with only
one-sense, element body, 25
26, 138, 161, 161n, 162-163 sthiti - duration of karma, 55 śubha-upayoga - auspicious manifestation of consciousness, 98, 106n, 108-109, 112, 114-116, 118120, 122, 133-134, 143, 145,
177, 181-182, 219, 225-228 śubha-yoga - virtuous activity,
49, 49n, 81, 138 śuddha-upayoga - pure manifestation of consciousness, 106, 108, 112-115, 117-119, 120n,
taṇhā - thirst, craving,
grasping, 38, 154 tapas - austerity, 20, 109, 153,
155-156, 186-188, 196-200, bāhya (external) listed 196197, uttara (internal) listed 197, 202-203, 222-223, 280, 285, 304, 306, 308 Tapassī, 12 tathāgatagarbha - Buddha
Nature, 282 Tatia, N., 150, 185 Tātparya-vrtti, 171, 231, 242,
258, 296 tattva - real, existent, 187, 199,
24, 294; svatattva, 214 Tattvadīpikā, 96, 113, 118-119,
121, 125, 128-129, 131-136, 138-139, 141-145, 148, 154, 157, 161, 169,-170, 172-174, 179, 186, 201-202, 204-205, 208-210, 213-215, 218-220,
222, 228 Tattvartha Sūtra, 2, 5, 11, 14,
36, 44-45, bondage in 46-90,
121, 120. 141-145, 172-174,

Page #368

354 Harmless Souls
91-92, 102-104, 107, 109110, 114, 116-117, 149-150, 152-156, 160, 168-169, 174, 182, 185, 187, 189, 192, 196-197, 200, 208, 220, 231,
291, 298; see also Umāsvāti Tattvārthādhigama Bhāsya
(Svopajna Bhāsya), 46-47, 47n, 102, 150 Terāpantha (Svetāmbara sub
sect), 183n-184n Thānamga Sutta, 49 three jewels (ratnatraya), 185,
187, 235, 298 trasa - mobile jīva with 2 or more senses, 6, 25-26, 73,
138, 161n, 161-163 two truths doctrine, 95-96, 232,
239-266; two 'two truths' doctrines, 254-260
140, 147, 154, 156-157, 160, 162, 167, 169, 175, 177, 209, 219-221, 224, 229, 231, 257, 267, 269, 289, 291, 305-306; development of doctrine, 97111; sāgāra and anāgārauvaoga, 100-101; see also darśana, jñāna, aśubhaupayoga, śubha-upayoga,
suddha-upayoga utsarga-samiti - care in
performing excretory
functions, 69-71 Uttarajjhayana Sutta, 4, 30.
49, 57, 97-99, 171 uvaoga - see upayoga
U
Umāsvāti, 2, 5, 38, 44, 46-90,
108, 110, 152, 175, 182, 196, 273, 294, 298; see also
Tattvārtha Sūtra upacāra - metaphor, figure of
speech, 243 upadātā - appropriator, 143 upadhi (uvadhi) - attachment to
possessions, 171-175 upādhika - limiting adjunct,
Vātsyāyana, 233 Vavahāra Pithikā, 19 Vedānta, 120, 126, 139, 141
142, 238, 243, 244n, 256, 265, 303, 305; Advaita
Vedānta, 137, 154 vihāra - monastery, temple,
298-299 vikalpa - false notion, 268 vinaya - rules of conduct for
Buddhist monks, 164 violence, 5-11, 32; see also
himsā, ārambha vīriya - see vīrya vīrya - will, energy, 99, 101 višuddhi - purification, 138 Viyāhapannatti (Bhagavai),
40-42, 44-45, 38, 86, 99-101,
104 volition, 13, 15, 22, 29-30, 54,
66, 273, 306; see also cetanā, vrata - vow, restraint, 84-86.
166, internalisation of 189, 226, 280, 308; see also anuvrata, mahāvrata
132
Upadhye, A.N., 91-96, 115
116, 157, 159, 168, 171, 179,
201, 218, 302 Upāli, 12 Upālisutta, 12-13 upāya - means, 251 upayoga (uvaoga) -
manifestation of consciousness, defining characteristic of the jīva, 97123, 129, 132-134, 138n,

Page #369

Glossary and Index
355
vratin - one who takes a vow,
85 Vyāsa, 207 vyavahāra-naya - conventional
view, 95, 109, 120, 122n, 123n, 126n, 130, 133-135, 137, 142, 144, 201-202, 206, 212, 224, 232, 239-266, 269, 272, 279, 297, 305
W
Williams, R., 190
Y yati - ascetic, 134 yogal-action, activity,
vibration of body, speech, and mind, 5-15, 20, 41, 49, 55, 59, 71, 75, 83, 88, 107, 109, 145, 152, 185, 204, 219-220; defined in the
Tattvārtha Sūtra 47-51, 49n yoga2- meditative exercise,
207 Yoga Bhāsya, 47, 207 Yoga Sūtra, 47, 52, 207 Yogindu, 302-303

Page #370

Page #371

Will Johnson was educated at the University of Sussex and Wolfson College, Oxford, where he was Michael Coulson Research Fellow in Indology from 1991 to 1992. He is now Lecturer in Religious Studies at the University of Wales, Cardiff. He is currently working on a book on Jaina art, and a translation of the Sauptikaparvan of the Mahābhārata. He has recently published a new translation of the Bhagavad Gītā.
21 -8.98 2421147 SA

Page #372
Lala S. L. Jain Research Series . General Editor: Prof. Dayanand Bhargava Vol. I: Jaina Yoga: A Survey of the Mediaeval Sravakacaras-R. Williams Vol. II: A Comparative Study of the Jaina Theories of Reality and Knowledge_Y.J. Padmarajiah, Vol. III: Dravya-Samgraha of Nemichandra Siddhanta Chakravarti-Ed. S.C. Ghoshal Vol. IV: The Essentials of Bhagavan Mahavira's Phi losophy: Ganadharavada --Acharya Vijay Bhuvanbhanusuri Vol. V: The Scientific Foundations of Jainism -KV. Mardia Vol. VI: Lord Mahavira and His Times - Kailash Chand Jain Vol. VII: The Jaina Theory of Perception -Pushpa Bothra Vol. VIII: The Jaina Iconography --B.C. Bhattacharya Vol. IX: Harmless Souls-Karmic Bondage and Religious Change in Early Jainism with Special Reference to Umasvati and Kundakunda -W.J. Johnson MOTILAL BANARSIDASS PUBLISHERS PRIVATE LIMITED